CPI ... Your Confidence Level in this Statistic?

[Gone4Good]

Some more data:

According to the Energy Information Agency,

Average residential electricity cost 2.6 cents per kilowatt hour in 1960 and 8.94 cents in 2004.  If you adjust the 1960 cost of 2.6 cents using CPI, electricity should cost 16.6 cents in 2004. 

In 1976 a gallon of unleaded gasoline cost 60.5 cents on average.  In March of 2006 the average cost for a gallon as 231 cents.  If you adjust the 1976 price of 60.5 cents using CPI, a gallon of gasoline should cost 216 cents in 2006.


Based on some spot checking that I've posted here (and in an earlier post), it does seem as if CPI is a pretty good barometer of prices.  In some cases its unreasonably low (as in tracking automobile prices) and in some cases it is unreasonably high (as in tracking electricity prices) - but I guess that is to be expected in any broad average. 

The idea that CPI is wholly unrepresentative of changes in prices seems not to be supported by the data.

 

[Gone4Good]

No, it is not.  The naive investor who reads 'real return' feels that this produces an inflation neutral return.  It is not that.

Hmmmm.  Is it true that everyone's personal rate of inflation is higher than CPI?

 

[d]

The naive investor who reads 'real return' feels that this produces an inflation neutral return. It is not that.

By definition, I think it is that; however, it is not likely a cost-of-living neutral return.  A "real" is simply a "nominal" relative to some other "nominal" ... one could use whatever they wanted for the reference, but it is commonplace to use CPI as the reference.

Hmmmm.  Is it true that everyone's personal rate of inflation is higher than CPI?

Providing the components reasonably represent what everyone (in aggregate) is purchasing, it's not possible ... for some it will overstate, for others it will understate. As the components are considered fixed during the period being measured, CPI tends to understate on the whole.

 

[haha]

As the components are considered fixed during the period being measured, CPI tends to understate on the whole.

Or so the Germans would have us believe.

Ha

 

[Nords]

Some more data:
According to the Energy Information Agency,
Average residential electricity cost 2.6 cents per kilowatt hour in 1960 and 8.94 cents in 2004. If you adjust the 1960 cost of 2.6 cents using CPI, electricity should cost 16.6 cents in 2004.
In 1976 a gallon of unleaded gasoline cost 60.5 cents on average. In March of 2006 the average cost for a gallon as 231 cents. If you adjust the 1976 price of 60.5 cents using CPI, a gallon of gasoline should cost 216 cents in 2006.
Based on some spot checking that I've posted here (and in an earlier post), it does seem as if CPI is a pretty good barometer of prices. In some cases its unreasonably low (as in tracking automobile prices) and in some cases it is unreasonably high (as in tracking electricity prices) - but I guess that is to be expected in any broad average.
The idea that CPI is wholly unrepresentative of changes in prices seems not to be supported by the data.

This electricity model is no better than a Model T.

Utilities have hedonics adjustments too. I grew up in an electrical-engineering family (father & grandfather) and I used to hear this "too cheap to meter" crap all the time. Electricity in the 1950-60s was created from burning oil & high-sulfur coal, or from nuclear plants with no realistic provisions for refueling or nuclear-waste disposal. No one was making utilities use clean fuel or stack scrubbers or worry about trading CO2 credits. Electricity was much dirtier then (variations in both voltage & frequency) and power reliability was much lower (overloaded poles, no tornado/hurricane resistance) with much more manual switching. The grid wasn't computer controlled across the entire TVA from coast to coast and up into Canada.

Same hedonics with gasoline. Hawaii just passed a legislative mandate to make its "gasoline" at least 10% ethanol. California varies its fuel mix with the seasons as well as all sorts of other additives. "Leaded" gasoline is a thing of the past and most of the additives in use in the '70s are considered to be too polluting today. BTW my fillup at a "cheap" station yesterday was $2.81/gallon and most of Oahu is over $3/gallon.

 

[mathjak107]

just reviewing our expenses in 2005 i cant come up with much that isnt up more than 3%..........my auto insurance stayed the same i see but every thing else is way higher....i think my general expenses are probley closer to 5% here in new york....couple that with no raises this year at work and it hurts.....

 

[Focu$ed]

This is always a fascinating topic. I have skimmed through the replies and read the interesting article at the link provided by CFB. My personal view is that inflation does appear to be higher than reported, and we personally are doing whatever we can to lock in some of our expenses at today's prices, or reduce our exposure to price increases. Not easy to do, unfortunately. The main thought I always come away with when thinking about inflation is that it seems unfair that we pay taxes on the "increase" in our investments when the increase is more or less a function of money being devalued by the increase of government debt and increased money supply, raising the price of everything, including our investments.

 

[Gone4Good]

This electricity model is no better than a Model T.

Utilities have hedonics adjustments too.  I grew up in an electrical-engineering family (father & grandfather) and I used to hear this "too cheap to meter" crap all the time.  Electricity in the 1950-60s was created from burning oil & high-sulfur coal, or from nuclear plants with no realistic provisions for refueling or nuclear-waste disposal.  No one was making utilities use clean fuel or stack scrubbers or worry about trading CO2 credits.  Electricity was much dirtier then (variations in both voltage & frequency) and power reliability was much lower (overloaded poles, no tornado/hurricane resistance) with much more manual switching.  The grid wasn't computer controlled across the entire TVA from coast to coast and up into Canada.

Same hedonics with gasoline.  Hawaii just passed a legislative mandate to make its "gasoline" at least 10% ethanol.  California varies its fuel mix with the seasons as well as all sorts of other additives.  "Leaded" gasoline is a thing of the past and most of the additives in use in the '70s are considered to be too polluting today.  BTW my fillup at a "cheap" station yesterday was $2.81/gallon and most of Oahu is over $3/gallon.

Nords, I don't think this is the right analysis for the data I put up. According to EIA the nominal average cost for electricity and unleaded gasoline is what I showed in the post above. I then used the change in the CPI index (not a component of CPI but the whole index) to calculate what CPI says those things should cost today and compared that cost against what things really do cost today. In the case of electricy, it has increased at a far slower pace than CPI and gasoline has increased at a slightly higher rate.

 

[cute fuzzy bunny]

Hmmmm. Is it true that everyone's personal rate of inflation is higher than CPI?

No it isnt, but it IS true that people who dont read the whole thread fail to notice that I said very clearly, several times, that some peoples personal rates are lower and some are higher.

By definition, I think it is that; however, it is not likely a cost-of-living neutral return.

It is CPI neutral (not inflation neutral), to the extent that CPI measures inflation, and it is a general/average neutral, not a regional or individual specific neutral. It is definitely not a cost of living neutral. Way too many differences in peoples COL to ever do that.

If anyone is still champing at the bit to "prove me wrong", we're only disagreeing if you agree with the following statement.

"CPI = inflation, and the purchase of "inflation protected" government securities or CPI indexed bonds gives me an inflation neutral investment that produces a predictable and real return that allows me to completely or nearly completely remove inflations effects from my long term retirement planning. For everyone else too."

If you agree with the following statement, then we're in agreement.

"CPI is not exactly equal to inflation for most people. Some will find CPI to be a higher measure than their personal rate of inflation, some lower, a very small number exact. CPI indexed securities are good ideas in times of very high inflation because few other options would pay as well as a CPI security should we see double digit inflation. CPI indexed securities at coupons below 3% for tips and 2% for ibonds are not very good investments when the CPI is in the low single digits. In any case, the ownership of these securities does not full eliminate the factoring of inflation into your personal retirement planning."

My big two points are that CPI <> inflation (and I dont hear a lot of disagreement on that), and that "inflation protected" investment products are perhaps a bit misleading in that they do not truly protect each investor from the effects of inflation.

I thought about this a lot more yesterday and the upshot of it is that I wonder if inflation doesnt have a far more profound effect on the ER than the average person. Heck, the hit from healthcare alone might be a bigger bite than the 3-5% of your total annual budget that CPI claims is the inflation rate.

The odd thing is that for me, inflation should have a lower bite. We dont eat out much, I dont pay for many services choosing to do most maintenance and repairs myself. Healthcare is paid through my wifes part time job.

Yet when I take our year on year spending, cull out unusual expenditures (like the sudden appearance of these mysterious "baby expenses") until I get it to basic core living costs...i'm seeing approximately a 6% growth year on year over the past 3 years. We havent made any conscious decisions to live a different lifestyle nor do I feel like we've 'turned up the knob' in any significant manner. Further, as I've outlined in posts on "invisible inflation",I see quality and quantity reductions regularly in the products and services that I buy. Customer service has been reduced to a joke at most companies. Food portions are smaller. "Extras" you used to get for free now cost extra.

For some, perhaps a lot of people, particularly living away from the coasts and in more rural, low-growth areas and living simply might be see CPI as extravagant. In my area, more urban, near the california coast, in a high growth area...I think CPI is grossly understating the effects of inflation.

I think this bears the ER or prospective ER needing to very closely examine their personal rates of inflation and not just relying on an average or a passing plug of the CPI into the retirement calculator.

To heck with stock market valuations, average rates of return, portfolio diversification and all that. It isnt going to mean squat if you're using a 3-3.5% long term average of CPI in your 40 year calculations and you're off by 2-3% a year.

Up until this point, I was ready to flag on anyone with a seven figure portfolio and no debt to zap their jobs and move forward to ER. Now I think a closer look at the influence of inflation on their lifestyles, long term, particularly around some highly ER enhanced items like health care and educational costs for ER parents, is very well warranted.

In fact, excepting people living away from high growth areas near the coast and people with pretty large sized portfolios, I'm thinking you better have at least a small income stream of some kind in place. Investment gains from a reasonable portfolio simply may not be enough.

As far as cherry picking a few items and looking at their relative cost factors, make sure you look at housing, college costs and health care costs before bothering to dip into piddlers like gas and electricity. Both of which have been kept low by artificially low oil and coal prices.

Hell...even the six million dollar man turned into the six billion dollar man in just 30 years...
http://www.guardian.co.uk/science/story/0,,1754356,00.html

 

[Gone4Good]

just reviewing our expenses in 2005 i cant come up with much that isnt up more than 3%..........my auto insurance stayed the same i see but every thing else is way higher....i think my general expenses are probley closer to 5% here in new york....couple that with no raises this year at work and it hurts.....

Unless you are tracking how many ounces of toothpaste you consumer from one year to the next and how many pounds of chicken you eat relative to how many pounds of beef, etc. I don't think those kind of comparisons are worth very much.  

I'm on track to spend about 15% less this year than last year . . . deflation?

Is it possible that this presumed mismatch between CPI and "personal inflation" is a result of the fact that almost everyone is consuming more?  Has gasoline outpaced CPI?  Not since 1976 it hasn't.  But people may very well be spending more on gasoline because they are driving more and instead of owning one 4 cylinder hatchback they now have two SUVs parked in the driveway.  Electricity prices have fallen in real terms but people may be spending more because they are consuming more electricity to run air conditioners that prior generations didn't have to cool houses that are 50% larger on average.

What you are seeing is not a mismatch between CPI and inflation, but keeping up with the Joneses run amok.  It's been noted elsewhere on this site that one of the hazards of matching CPI is that your standard of living remains constant in a world where working stiff’s standard of living is increasing – which makes you feel like you are losing purchasing power in comparison to your neighbors.  That is not a problem with the price index.  The index seems to accurately track prices.  What it does not track is everyone’s insatiable appetite to consume more.  

 

[cute fuzzy bunny]

But I live in the same house, drive the same cars, am not comparing my costs with prior generations, and i'm far from the "keep up with the jones's" lifestyle guy.

My consumption, year on year, is pretty comparable from what I can see.

I guess the big question for me is why you seem a bit hell-bent on defending the CPI?

When I read the hedonics paper on electronics and it made the most half-assed pitch for why computer costs should be ratcheted down because they simplify our lives by being faster than last years crop (which is bullshit), and after my third read of how health care cost changes are applied and came to the conclusion that they really arent (which is bullshit), and noting that CPI only factors in the fairly stable rents and not home purchase costs (which is bullshit), I stopped regarding the statistic as being very effective in measuring inflation.

 

[wabmester]

I thought about this a lot more yesterday and the upshot of it is that I wonder if inflation doesnt have a far more profound effect on the ER than the average person.  Heck, the hit from healthcare alone might be a bigger bite than the 3-5% of your total annual budget that CPI claims is the inflation rate.

Almost certainly. In fact, the BLS came up with an experimental CPI for retirees (CPI-E) a while back, and it looks like the CPI-E was generally about half a point higher than CPI-U. And these were retirees with medicare benefits, not ERs who pay their own way.

 

[Nords]

...an experimental CPI for retirees (CPI-E) a while back, and it looks like the CPI-E was generally about half a point higher than CPI-U.    And these were retirees with medicare benefits, not ERs who pay their own way.

They must've decided that:
- they were scaring people off, or
- they were going to have to address the issue of retirement CPI-adjusted investing, or
- the whole project was irrelevant if these people weren't planning to or saving enough to retire in the first place.

 

[cute fuzzy bunny]

http://www.aarp.org/research/socialsecurity/reform/aresearch-import-327-DD52.html

Interesting read on the CPI-E.

Most of the difference in the CPI-E being higher is higher medical costs and slightly higher housing costs (?), the rest of the costs are dialed downwards.

 

[d]

I, for one, can certainly agree that CPI will not accurately reflect "individual inflation" rates for everyone; and can further agree that "inflation protected securities" do not securly protect against inflation. But I'd not, as some (no, not you C.F.B.) seem to, jump to the conclusion that it's a result of some vast conspiracy.

Two issues were raised in the more recent posts which are certainly deserving of attention: 1) tax impacts and 2) the housing component.  

Among the reasons that "inflation protected securities" do not securly protect against inflation is that we pay taxes on the return, so that the after tax return is clearly less than "advertised" and if inflation itself is large enough, the "real" after tax return will indeed be negative.  

CPI measures the housing component by including the rental value of owned property, i.e. not the current price of residential realestate.  This is a reasonable approach, since at least in part home ownership is investment rather than consumption; but the result is that over short periods of time CPI does not reflect complete home ownership costs.

 

[cute fuzzy bunny]

I dont hear any "vast conspiracy" arguments, although I cant say i've ever been comfortable with the idea of the govt figuring out how successful their economic policies are and making payments against something when they measure it themselves. Consciously or subconsciously, I find it tough to believe there isnt any influence or effect to that, albeit minimal.

I honestly think they overthink the whole thing, and that the overthinking produces a somewhat bogus result. If they dropped the hedonics and the basket substitutions, I'd probably feel a lot better about it. There is an opportunity to morph cost of living/inflation measurement into an adjusted quality of life measurement. That aint supposed to be what its for.

CPI and housing? Try that trick in california where you can rent a house for $1000-2000 a month that costs $400k-800k to buy. Housing prices have been leaping almost every year for over a decade. Rents...not so much. If a decade == "short periods" per your description...well...okay. Rents tend to be tied to how much a paycheck-to-paycheck renter can afford. Housing prices are a different beast. In some areas of the country. Not a problem for a homeowner thats indexed into the current high prices. Good thing for someone who can sell a 400k house and buy the same home somewhere else in the country for $200k. Bad problem for someone wanting to move into the area or for a new buyer thats been renting.

Understand that I'm prewired for 440V on the "real return" thing because we had a pair of trolls running around here a little over a year ago with the hairbrained idea to buy a 100% tips portfolio, take the "2.5% real" payout, and eat 1/40th of the portfolio principal every year, over a 40 year period. The "completely safe 40 year ER" was the pitch. But it ignored the fact that the "real" return wasnt "real" for everyone, and the fact that turning off the life switch after 40 years might not be too thrilling for a lot of people.

 

[d]

..."real" return wasnt "real" for everyone

totally agree

 

[sgeeeee]

This is the best discussion we've had yet on inflation and CPI. Someone should sticky it or bookmark it for the next time the question comes up (about once every other month apparently).

 

[mathjak107]

im going back to my slice of pizza at the local pizza shop index

 

[d]

im going back to my slice of pizza at the local pizza shop index

to each his own

 

[Gone4Good]

I guess the big question for me is why you seem a bit hell-bent on defending the CPI?

I actually came to this discussion pretty much as an agnostic on the issue. I will admit my predisposition was to think the government stats were pretty good. I never believed they were massively manipulated as some people seem to think. I did assume there was a tracking error, but judging by the discussion on this board, the tracking error was huge. By some people's count (and I think CFB counts among this number) the tracking error was somewhere in the 3% range. So I wanted to look at historic prices where I could find them and see how badly CPI tracked with what we know prices are today. And low-and-behold what I found was that CPI tracks pretty darn well - it certainly doesn't seem to be off by anywhere near 3% per annum.

Now as hard as I look in the 5 pages of posts, no one has claimed my numbers were bad or posted alternative calcs that show a different result. Notwithstanding that, CFB and others are "hell-bent" on proclaiming CPI "is a joke."

It's a price index! And from what I can tell it seems to do a pretty good job of tracking prices over pretty long periods of time. If you want it to do something else (which seems to be CFBs real gripe) then you won't get any arguments from me. It is not, nor does it claim to be, an individual consumption index.

 

[d]

well said

 

[Gone4Good]

But I live in the same house, drive the same cars, am not comparing my costs with prior generations, and i'm far from the "keep up with the jones's" lifestyle guy.

I don't consider myself a keeping up with the jones's kind of guy either, but when I look back over the past 15 years here are some of the things that factor into my "personal inflation rate":

1) 15 years ago I didn't own a computer . . . now we own three.  Along with the expense of the three computers comes a printer, toner, paper, wireless router, etc.

2) No internet connection 15 yrs ago . . . then dial-up service . . . now a cable modem

3) 15 years ago we had a basic land-line telephone connection . . . now our land line has call waiting, caller id, etc. all of which costs extra

4) 15 years ago we didn't have A/C . . . now we have central air

5) 15 years ago we had basic cable service with about 30 channels . . . now we have digital cable with more channels than I can count plus video on demand

6) 15 years ago we didn't have cell phones . . .  now we have two

7) 15 years ago a vacation was a road trip to Cape Cod . . . now it is a flight to Europe

8 ) 15 years ago we had a VCR . . . now we have a DVD player and Tivo

All of this stuff, and more, comes about incrementally.  You don't think you are spending that much more when you first sign up for the cell phone, or add telephone services, etc. but it adds up to a lot of dough over time.  That doesn't even consider the incremental costs of "premium" brand purchases (e.g. Starbucks vs. Maxwell House, wine instead of beer). 

Obviously my "personal inflation rate" over the past 15 years is greater than CPI . . . but that is not because CPI failed to accurately track prices.

 

[cute fuzzy bunny]

Now as hard as I look in the 5 pages of posts, no one has claimed my numbers were bad or posted alternative calcs that show a different result.

Look again. I posted a web site that offers a very reasonable alternative set of numbers. I also posted my own personal numbers, which are different.

Notwithstanding that, CFB and others are "hell-bent" on proclaiming CPI "is a joke."

Never said that. Not even once. Not even anything close to that.

I even spelled out exactly what I agreed and disagreed with. Never said there was a conspiracy. Never said it was a "joke".

Now actually read what I said and tell me where we diverge.

I'll save you the time: I am "hell bent" on making sure people understand that CPI<>inflation, and that "inflation protected" pensions, social security, investment bonds and other CPI indexed products are not "inflation protected" to their cost of living (necessarily) and that early retirees may in some instances be more succeptible to inflation than wage earners.

No more. No less.

It's a price index! And from what I can tell it seems to do a pretty good job of tracking prices over pretty long periods of time. If you want it to do something else (which seems to be CFBs real gripe) then you won't get any arguments from me. It is not, nor does it claim to be, an individual consumption index.

It is utilized as a broad based indicator of inflation, which I'm not sure its very good at due to the lack of exposure to housing prices, medical insurance and other major factors that effect peoples out of pocket expenses quite strongly. It is used to apply cost of living increases to pensions, social security and "inflation protected" investments.

 

[Nords]

It's a price index! And from what I can tell it seems to do a pretty good job of tracking prices over pretty long periods of time. If you want it to do something else (which seems to be CFBs real gripe) then you won't get any arguments from me. It is not, nor does it claim to be, an individual consumption index.

It may be great at tracking prices in the aggregate, but it's not so good at applying them to the individual. Unfortunately that's where our inflation-adjusted pensions & investments collide with the government's irresistible desire to mess with the numbers. Hedonic adjustments are especially slippery.

It also does not track well over long periods of times. Read "Triumph Of The Optimists" and their "adjustments" to a century of inflation data from indices including whalebone corsets... and that's the American CPI, let alone those of 15 other countries.