PDA

View Full Version : Consumer spending = inflation?



c1ue
07-01-07, 08:17 AM
Here's a possibly stupid question:

Consumer spending has been increasing steadily for the last 7 years despite real wages not rising - netting us several years of negative savings rates.

In 2000 the rising stock market was touted as the reason, later followed by the theory that home equity was rising, but neither excuse is valid any longer.

My question is: Given that the apparent inflation rate seems higher than the official rate, how much of consumer spending is simply reflecting 'actual' inflation?

Some excerpted numbers (from bls.gov, inflation date using calculator)

Year Overall Spending Food Housing Transportation Official Inflation
2000 2.0____________2.5__2.2____5.8__________3.361
2001 3.9____________3.2__5.6____2.9__________2.846
2002 2.9____________1.0__2.1____1.7__________1.581
2003 0.3____________1.0__1.1____0.3__________2.279
2004 6.3____________8.3__3.6____0.3__________2.663
2005 6.9____________2.6__9.0____7.0__________3.388

Note that Housing, Transportation, and Food are the 3 largest spending categories - in that order.

If I weight each of these 3 categories by their share of overall spending and multiply by the increase in the category, the 'FHT' contribution to consumer spending growth looks like:

Year Overall Spending Food__Housing Transportation Official Inflation 'FHT' Inflation Avg. Gas Price
2000 $38,045_______$5,158 $12,319 $7,417________3.361________2.182________$1.551
2001 $39,518_______$5,321 $13,011 $7,644________2.846________2.835________$1.421
2002 $40,677_______$5,375 $13,283 $7,759________1.581________1.142________$1.397
2003 $40,817_______$5,340 $13,432 $7,781________2.279________0.550________$1.513
2004 $43,395_______$5,781 $13,918 $7,801________2.663________2.314________$1.911
2005 $46,409_______$5,931 $15,167 $8,344________3.388________4.532________$2.290
2006 $2.981
2007 $3.146*

I throw in the average gas price figure for contrast - these numbers are from July of the respective year except for 2007. The 2007 number is for May (latest available from http://tonto.eia.doe.gov/oog/info/gdu/gaspump.html)

Assuming I am not making some really stupid mistake (it is 5 am my time as I write this), this data would seem to imply that 'FHT' spending was trending down from up to 2003 despite a spike in 2001 (due to real estate liquidity diversion from stock market?), but has been trending up since then.

More importantly, 2005 shows 'FHT' spending significantly greater than official inflation - more than 1/3 greater!

Transportation numbers are more or less consistent with average gas price.

Housing expenditure figures do not reflect housing appreciation until 2005, but the ARM/IO loans could explain that.

Food shows an inexplicable large jump in 2004, but perhaps gas prices explained that.

I am very eagerly awaiting the 2006 data...

bart
07-01-07, 04:15 PM
...
My question is: Given that the apparent inflation rate seems higher than the official rate, how much of consumer spending is simply reflecting 'actual' inflation?
...

Here's one view. The black line is actual retail sales, the blue shows it adjusted by CPI and the green shows it adjusted by CPI + lies (lies adjustment from shadowstats.com plus my own judgments).

As you can see, retail sales has basically been flat since about 2006 with just a CPI adjustment, and declining since 2000 when including the lies.


http://www.nowandfutures.com/images/retail_sales_cpi_lies1992-current.png

lb
07-03-07, 04:56 PM
Excellent graphing as always.

I presume the same thing would apply to this data:
http://www.census.gov/indicator/www/m3/index.htm

Where the increase in nondurable goods overshadows the decrease in durable goods (as measured by bonars/d0llars), just enough to say we have a healthy economy.

Nondurable goods includes non core CPI data such as food and gas...

bart
07-03-07, 05:49 PM
Excellent graphing as always.

I presume the same thing would apply to this data:
http://www.census.gov/indicator/www/m3/index.htm

Where the increase in nondurable goods overshadows the decrease in durable goods (as measured by bonars/d0llars), just enough to say we have a healthy economy.

Nondurable goods includes non core CPI data such as food and gas...

Thanks! :) I am working on making my various charts easier to understand too, especially the ones with many lines.

I don't track much in manufacturing but did add non durables few months ago to help see more of actual consumer spending patterns. Here's the short and long term pictures - flat this year with just a CPI adjustment and down when lies are added in.

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



http://www.nowandfutures.com/images/non_durables_cpi_lies1963-current.png

lb
07-03-07, 06:37 PM
My god, TMG is right, We Are Doomed!

bart
07-03-07, 06:48 PM
My god, TMG is right, We Are Doomed!

:D

... and that's *Freakin'* Doomed... ;)

c1ue
07-04-07, 06:42 AM
Bart,

I was wondering - your graphs are great but they seem to show divergence between CPI (official) and CPI (adjusted) starting around 1984.

The implication is that there was some muddling of CPI starting in the 2nd term of Reagan, significant escalation in the 2nd term of Clinton, followed by an initial levelling off then return to escalation in Bush II term 2.

When I get the chance, I will update my graph back to 1980.

One item that was interesting in my short little exercise was the relative impact of food, housing, and transportation on the overall inflation rate.

These 3 categories add up to around 2/3rds of overall spending - thus any manipulation of these categories either has to involve FHT directly or must involve outsize manipulation of one or more of the remaining categories in order to have an effect.

Whenever sleuthing manipulation of numbers - it may be possible to look for inconsistencies within the data itself since full consistency is much more difficult to achieve vs. a target output number.

bart
07-04-07, 11:56 AM
Bart,

I was wondering - your graphs are great but they seem to show divergence between CPI (official) and CPI (adjusted) starting around 1984.

The implication is that there was some muddling of CPI starting in the 2nd term of Reagan, significant escalation in the 2nd term of Clinton, followed by an initial leveling off then return to escalation in Bush II term 2.
...


Exactly true on the CPI fiddling having started in the early '80s. That's when the Homeowners Equivalent Rent game started at the BLS, where they don't use the actual housing prices but rather an odd fiction based on the concept of the home owner renting the house back to himself.

The mid '90s is when both geometric weighting and the results of the Boskin Commission idiocy were added in. Substitution bias was part of the Boskin package, and in plain English it allows things chicken to be substituted for beef when beef goes up and chicken stays the same or goes down. In other words, substitution bias makes the CPI no longer useful for doing true apples to apples comparisons over time.

Hedonics also came in during the early part of this decade.
From my glossary:

Literally it means relating to pleasure. But in an economic sense in US statistics, it involves adding or subtracting values to important government statistics like CPI or GDP that cause them to be false or incorrect.

One general example is in the computer area. If you bought a $2000 computer 3 years ago, and then replaced it with another $2000 computer today that is twice as fast, it can be counted as roughly a $4000 computer in the GDP since its twice as fast. This completely ignores that a computer that is twice is fast does not double the amount of work done by or with that computer, or even close.
According to Pimco’s Bill Gross, the US Bureau of Labor Statistics has expanded the use of hedonic adjustments and applies these adjustments to everything from computers, DVDs, automobiles, washers/dryers/refrigerators to college textbooks. Hedonics is used to adjust as much as 46% or more of the weight of CPI.
...
Using BLS statistics, health care costs are about 17.5% of consumption, but it is weighted much less in the CPI calculation. Healthcare is 4.6% of CPI; healthcare commodities are 1.5% of CPI. Healthcare is reportedly 15 to 17% of GDP. This presents a huge discrepancy in CPI weighting. If CPI healthcare costs were in tune with reality AND they had an accurate weighting, CPI would be substantially greater
...
Another issue with hedonically adjusting for better or faster products is that the reverse is not done. In other words, if the quality of something drops over the years no adjustment is made for it. One example would be solid wood furniture versus using a thin veneer.



For much more detail and data, visit http://www.shadowstats.com and read John Williams excellent work in the whole area. He's the primary source for the adjustments I use in my charts.

DemonD
07-07-07, 05:13 PM
clue, if you are just looking at the total numbers, and not number per capita, then you might not be seeing the whole picture. If US population grows by 2.5%/year, and consumer spending goes up 2.5%.... well there ya go. Also decreasing unemployment would lead to larger consumer spending even if the median or mean salary is the same.

incidentally i did a cursory search on us popluation growth but couldn't find anything useful for this post. i would admit population growth wouldn't answer the entire question but that it would definitely make up a part of total consumer spending.

c1ue
07-10-07, 02:47 PM
DD,

Consensus on US population growth is around 1.3% with children per woman at around 2.1

Wikipedia: US population growth in 20th century: 1.3% (probably includes immigration)

From census.gov:

US population in 2000 - 281421906
US population in 1990 - 248709873
Using the (x)exp10*Pop1990 = Pop2000 thus (x)exp10 = Pop2000/Pop1990 and x = e(ln(Pop2000/Pop1990)/10)

x= 1.0124

Thus according to the US Census, population growth from 1990 to 2000 was 1.24% - including immigration.

While this is non-zero - it does show that any price increases over 1.24%/annum is indeed inflationary.

Note that it is a common mistake to confuse the children/woman stat (2.1) with growth rate. This number implies a lower growth rate than 1.24%; the remainder is immigration.

c1ue
07-10-07, 02:53 PM
DD,

One other note:


Also decreasing unemployment would lead to larger consumer spending even if the median or mean salary is the same.


I understand where you are going, but you must know this statement is at face value nonsensical.

Food, Housing, and Transportation are 3 categories which I would posit would be the least affected by employment. If you've ever tried to get a job without a car or a place to live - I think you'd realize this :)

As for food, certainly discretionary spending can increase with increasing income. However, the minimum value is still significant w/ relation to average wages. Note that food spending is tabulated at around $5500 - $6000 per year, that is only $500 a month.

I don't know about you - but I'm a cheap bastard with regards to food spending and I semi-regularly exceed my $1000/month budget for myself and my wife. Even after excluding the $500/month we spend on eating out and drinking :o, the remainder is still significant even though I achieve 40% discounts on my Safeway bill 90% of the time.

Jim Nickerson
07-10-07, 11:05 PM
DD
I don't know about you - but I'm a cheap bastard with regards to food spending and I semi-regularly exceed my $1000/month budget for myself and my wife. Even after excluding the $500/month we spend on eating out and drinking :o, the remainder is still significant even though I achieve 40% discounts on my Safeway bill 90% of the time.

c1ue,

In the past 3 years, my wife and I have spent average of $295/mo for all groceries and dining out, and on no days do we eat little or poorly. I would suggest you don't know the meaning of "cheap bastard." It strikes me that two people spending 12K a year eating in and/or out offers some possible insight into why the US saving rate is negative.

c1ue
07-11-07, 12:33 AM
Jim,

I used to live in Austin - at that time and place (1996-2000) I was able to save over 50% of my pre-tax income while simultaneously buying a (resale) house and a new car (only one I've bought ever).

I am in tech - but I started out at $18/hr in 1996, reasonably darn good for a college grad but hardly Richie Rich.

However, I now live in the Bay Area.

Where once I spent $100/month on food, now I spend $500/month for just groceries. Note I don't shop at Whole Foods or any such - I only buy what is on sale and shop around at 3 different places.

However, while rent in Austin in 1996 was $500/month for a relatively top of the line apartment, my rent now is many multiples of that. I can only assume commercial real estate is similar and no doubt this drives costs up in stores.

Thus while I applaud your ability to thrive on $295/month in Fort Worth, I find I cannot do that here in the Bay Area.

As for the other $500/month - that is just 1 really good or 2 decent nights going out for a nice dinner and drinks.

You might also note that the average income in Fort Worth (according to Wikipedia) $37074 while the average income in San Francisco (where I live) is $57833.

Thus the spending differential is not nearly as dramatic as you think in terms of percentage of income.

But of course, I have state income tax plus a higher nominal tax rate.

Were I to own a home, no doubt I would also pay more property tax.

According to homegain.com, average Fort Worth house price is $145500.

Given the approx. 3% property tax rate I saw in Austin and assuming something similar for Fort Worth, that'd be around $4000/year.

The average house price in San Francisco is somewhere north of $600K. According to CNN - it is $656K. That'd net me property taxes (1.135% rate) of $7400+ a year.

Just bear in mind differentials in local markets when accepting dollar values!

Jim Nickerson
07-11-07, 10:13 AM
Jim,

I used to live in Austin - at that time and place (1996-2000) I was able to save over 50% of my pre-tax income while simultaneously buying a (resale) house and a new car (only one I've bought ever).

Very commendable, c1ue.



I am in tech - but I started out at $18/hr in 1996, reasonably darn good for a college grad but hardly Richie Rich.

However, I now live in the Bay Area.

Where once I spent $100/month on food, now I spend $500/month for just groceries. Note I don't shop at Whole Foods or any such - I only buy what is on sale and shop around at 3 different places.

However, while rent in Austin in 1996 was $500/month for a relatively top of the line apartment, my rent now is many multiples of that. I can only assume commercial real estate is similar and no doubt this drives costs up in stores.

Thus while I applaud your ability to thrive on $295/month in Fort Worth, I find I cannot do that here in the Bay Area.

As for the other $500/month - that is just 1 really good or 2 decent nights going out for a nice dinner and drinks.

You might also note that the average income in Fort Worth (according to Wikipedia) $37074 while the average income in San Francisco (where I live) is $57833.

Thus the spending differential is not nearly as dramatic as you think in terms of percentage of income.

But of course, I have state income tax plus a higher nominal tax rate.

Were I to own a home, no doubt I would also pay more property tax.

According to homegain.com, average Fort Worth house price is $145500.

Given the approx. 3% property tax rate I saw in Austin and assuming something similar for Fort Worth, that'd be around $4000/year.

The average house price in San Francisco is somewhere north of $600K. According to CNN - it is $656K. That'd net me property taxes (1.135% rate) of $7400+ a year.

Just bear in mind differentials in local markets when accepting dollar values!

c1ue, I expected I was not being objective or fair by picking on you; however, you defended yourself well. Very good. I appreciate your research and the contrasts of which it made me aware.

There are a lot of guys who disclose in one or another of their contributions here things such as where they live and what they do (underlined above), but if one looks at their profile by clicking their names, one sees nothing. It sure helps me differentiate and remember who contributors are when they have been so thoughtful to put where they live, what they may do, what they are interested in, etc. all in their profiles.

There are some good restaurants in SF, but I honestly do not know how one can stomach dinners running 250 for two. I cannot enjoy such food, but that is just me.

c1ue
07-11-07, 11:08 AM
Jim,

Not to worry. I read your stuff too and respect your opinions, hence my desire to elucidate rather than flame :rolleyes: