Semi OT: inflation

Berkshire’s stock price rose 35% in the 13 months from end December 2020 to end January 2022.
But that’s in nominal terms.
CPI adjusted, it was a rise of only 25.2%.
Transient or not, for the moment we are in a stretch that inflation really matters when sitting in your counting house counting out your money.
A 10% gap in such a short time is quite a bit to shrug off.

The curve sure has changed.
For the 23 months from end 2019 to end Jan 2022, it rose overall at 4.29%. High but not earth shattering.
But that was a game of two halves, as the expression goes.
From end December 2019 to end December 2020, CPI rose 1.28%.
In the next 13 months, it rose 7.79% (7.16%/yr rate).

I post partly because of how maddeningly difficult it is to find the right data source for the index level (281.933 for end January).
The BLS seems determined to make it hard to find anything other than 12 month year-on-year figures…the actual index level is very well buried.
I find it easiest to get from the St Louis Fed, here https://fred.stlouisfed.org/series/CPIAUCSL

And please, no anecdotes about how your personal expenditures have risen more, or less, than the
CPI-U figure in this stretch and therefore CPI must ipso facto be wildly wrong (way too high, or way too low).
It all depends what you’re spending on personally–every person’s expenditure basket is unique.
Overall, CPI-U is a pretty good metric of how much “stuff” a US dollar will buy: how well it’s holding its general purpose purchasing power.
As an investor, that’s the meaning I need to track my investments.
Though I do it in more than one currency.

Jim

13 Likes

Personal expediture choices do alter personal “inflation”…LOL. I just bought the white cedar siding and facia to make a cushioned reading nook 5 feet by 10 feet, about 5 feet in height. I bought the same siding I used on the house when it was built in 1985, a 3000 foot house with a full basement.

The tiny bit of facia and siding I bought this week cost more than all the siding on the original house in 1984.

5 Likes

Personal expediture choices do alter personal “inflation”…LOL. I just bought the white cedar siding and facia to make a cushioned reading nook 5 feet by 10 feet, about 5 feet in height. I bought the same siding I used on the house when it was built in 1985, a 3000 foot house with a full basement.

The tiny bit of facia and siding I bought this week cost more than all the siding on the original house in 1984.

When I had major remodeling done on my kitchen in 2006, it cost 3.8 times what the original owner paid for the whole house and lot when the house was built in 1950.

The Inflation Challenge
Navigating an inflationary environment

Buffett (1980): “High rates of inflation create a tax on capital that makes much corporate investment unwise - at least if measured by the criterion of a positive real investment return to owners. This “hurdle rate” the return on equity that must be achieved by a corporation in order to produce any real return for its individual owners - has increased dramatically in recent years. The average tax-paying investor is now running up a down escalator whose pace has accelerated to the point where his upward progress is nil.”

https://davidstokman.substack.com/p/the-inflation-challenge?..

3 Likes

When I had major remodeling done on my kitchen in 2006, it cost 3.8 times what the original owner paid for the whole house and lot when the house was built in 1950.

On the other hand, my old company sold a LOT of Kingston memory boards at $8500 per megabyte.
And we spent about $20,000 per month on telephone long distance charges.
Both of those cost less than a cent now.

Some pretty common things are close to flat in nominal terms. A lot of clothing, and some food.
In the spring 1980 Sears catalogue, a T-shirt was $3.99.
It’s now $5.39 at sears.com, 42 years later, up 0.72%/year.
That’s pretty flat, considering that CPI inflation in 1980 was running about 12-14% per year.
If you want really flat, you have to go to https://www.399tshirts.com/
Though it’s not really $3.99…they say you can take 15% off your entire order if you buy 3 or more products.

One’s personal basket will vary a lot–it depends what you like to buy, or have to buy.

Jim

11 Likes

“ Transient or not, for the moment we are in a stretch that inflation really matters when sitting in your counting house counting out your money.
A 10% gap in such a short time is quite a bit to shrug off.”

Still debating how much cash to hold. Currently 10% or so and bought some big tech last month (Google & Amazon) but not many fish in barrels to shoot right now. Inflation tax is painful but want some meaningful dry powder to use like 2020 and sure some Volatility will provide better opportunity. Fully anticipate WEB is still repurchasing in size but prefer a better discount than Mr. Market is offering.

Looks like WEB is keeping about 15% of total assets in cash and waiting patiently as he quietly continues the buybacks. Cannot wait for the annual report and Omaha- been a great 12 months across the board.

3 Likes

When I had major remodeling done on my kitchen in 2006, it cost 3.8 times what the original owner paid for the whole house and lot when the house was built in 1950.

On the other hand, my old company sold a LOT of Kingston memory boards at $8500 per megabyte.
And we spent about $20,000 per month on telephone long distance charges.
Both of those cost less than a cent now.

Very true. The first computer I ever used was an IBM 704 that cost us $680/hour to rent. You can now buy a PC for $680 or less that is far far more powerful. (I believe even my cell phone is more powerful) IIRC, the clock speed of that machine was about 1 MHz, and the memory required about 12 microseconds per cycle. You could get 32,768 words of RAM, maximum, for those machines. My main desktop has a 16-core processor that runs at 3.90GHz and has 64 Gigabytes of DDR4 RAM; and I could add another 64 Gigabytes if I had the money and needed it. The first hard drive I had was the size of a top-loading washing machine, ran at 2400 rpm, and held 40 megabytes. My hard drives on my current machine hold 4 TeraBytes each, spin around 7200 rpm, and they are not state-of-the-art. 10 years ago I had six 10,000 rpm SCSI hard drive in one of my machines.

Telephone bills are more difficult. IIRC, my first telephone bills, in the early 1960s, were well under $10/month, but that was voice-only and local calls only. My phone bill now (not including cell phone) is $112.68, but that includes long distance anywhere in USA (including Hawaii) and Canada, and also 75 Megabit/second fiber-optic Internet connection.

Cannot wait for the annual report and Omaha- been a great 12 months across the board.

Same here.
I wonder if the observable value per share, adjusted for inflation, will be back to its pre-Covid trend.
If so, they my wild guess from spring 2020 that there could be a full calendar year of lost progress might have been quite a bit too pessimistic.

Jim

6 Likes

mungofitch said:

I post partly because of how maddeningly difficult it is to find the right data source for the index level (281.933 for end January).
The BLS seems determined to make it hard to find anything other than 12 month year-on-year figures…the actual index level is very well buried.

BLS has a “most popular series” tool that will get you to CPI-U (and a few other series) fairly quickly:

https://data.bls.gov/cgi-bin/surveymost?bls

If you need more granularity (either geographically, by item(s), by weighting, etc.) this is the place to go:

https://www.bls.gov/data/

I think the “one screen” option is the best as it puts everything right in front of you.

Pro Tip: If you write down the Series ID for the data you want you can just search that at BLS…

For example, you appear to want the CPI-U seasonally adjusted index. Series ID: CUSR0000SA0. Searching for that at bls.gov, the top result is the data table. You might need to fiddle with the “formatting options” but it makes getting updates just a little bit quicker.

Also, the BLS series have some systematic naming to them. For example, the CPI-U not seasonally adjusted Series ID is CUUR0000SA0.

7 Likes

BLS has a “most popular series” tool that will get you to CPI-U (and a few other series) fairly quickly:

Sure, but searching a database seems like they are really trying to hide the index number.
The “I” stands for index!

And I still have a hard time finding it.
It’s not on the page you linked, for example.
The “usual” number is the “CPI for All Urban Consumers (CPI-U) 1982-84=100 (Seasonally Adjusted)”, CUSR0000SA0, which isn’t at that page.
If you search for it in the “data tools” / “one page” tool, it says “this database currently unavailable”.
You can find it at “data tools” / “prices - consumer” / “all urban consumers” / “All Urban Consumers (Current Series)” / “data finder” / then scan the list.
It’s the second one sorted by popularity.
Click on that to get a page with a graph, and the latest number is in a box at the bottom left of the screen.
You can see why I suspect they don’t actually want people to read the number : )

It appears nowhere in their 2991 word press release, yet pretty much everything most people want to know derives from seven characters: “281.933”
I’d have a badge on the home page with the current CPI-U index level, its effective date, and (yes) the trailing twelve month change.
Click on it and you’d get a page with a downloadable table of the monthly figures going back as long as you like.

Jim

9 Likes

I post partly because of how maddeningly difficult it is to find the right data source for the index level (281.933 for end January).
The BLS seems determined to make it hard to find anything other than 12 month year-on-year figures…the actual index level is very well buried.
I find it easiest to get from the St Louis Fed, here https://fred.stlouisfed.org/series/CPIAUCSL

Not that I blame you, but it is in fact here:
https://download.bls.gov/pub/time.series/cu/cu.data.1.AllIte…
The series to look at is CUUR0000SA0.
What could be simpler?

1 Like

What could be simpler?

Thanks. Nice sleuthing!

To your question: not simple enough for me.
It’s not on any of the pages I was ever able to find, or the ones linked in this thread.
And I’ve tried on many many occasions, though admittedly I do tend to give up in frustration quickly each time.
I’ve only ever succeeded in finding it at the Fed, so I’ve used that for a long time now.

Out of curiosity, how did you find that page?
By knowing the cryptic unique data sequence code ID and searching for that?
Or did you find a navigation path to it at the BLS site?

I remain convinced they are trying to make it hard to find.
It’s hard to believe it could be that difficult by accident–I’m reasonably bright : )

Jim

5 Likes

What could be simpler?

Thanks. Nice sleuthing!

I suppose it should be added that that file itself could hardly be much more cryptic.
Yes, technically it includes the figures I want.

But it contains ALL the CPI sub-index figures, not just the one most people want (all cities, usually seasonally adjusted)
It has 57 separate figures for January 2022 scattered through the file, with only cryptic code identifiers to distinguish them, no headers or metadata.
Without knowing in advance that what you want is called CUSR0000SA0, you’re out of luck.

Jim

7 Likes

Jim,

I honestly don’t remember how I stumbled on it. But I do remember getting frustrated (as I think you can empathize with). But that just made me more determined.

I tried “tables” and other such promising but fruitless leads, finally memorized the value of CPI-U as of Feb 2011 and went through the text files to see which one matched it for Feb 2011.

The most straightforward path to get there (which I did not take, and is still more crooked than a politician), is:
bls.gov > Subjects tab > CPI > All Urban current series > Text files > search through the non-specialized ones (i.e. ignore US housing etc.)

(Magic is always obvious once you explain it :-))

This is what a nerd like me does for thrill.

4 Likes

Transient or not, for the moment we are in a stretch that inflation really matters when sitting in your counting house counting out your money.
A 10% gap in such a short time is quite a bit to shrug off.

In Today’s Wall Street Journal

Big Companies Thrive During Periods of Inflation
https://www.wsj.com/video/big-companies-thrive-during-period…

It’s a video which I have not watched, but the headline caught my attention,

4 Likes

One word of caution.
Jim - is the series you need really the seasonally adjusted CPI-U (the CUSR…) one, or the unadjusted (CUUR…) one?
If you are using it to adjust earnings or sales or other financial statement figures, I would guess it is the latter. Either series would allow you to compare numbers from January 2012 to January 2017, for example. I am not quite sure what the interplay between the seasonality of earnings/sales etc and the seasonal adjustment for the CPI would produce. IOW what the better way is to compare earnings from the quarter ending in January 2017 vs the quarter ending in July 2020.

1 Like

what the better way is to compare earnings from the quarter ending in January 2017 vs the quarter ending in July 2020.

My opinion is that their seasonal adjustment method is pretty good, so that’s what I use.
When something truly dire happens, they don’t simply use the seasonal variation for the year before, for example.

So, personally, I do this:
Take the “most recently published and known at the time” CPI-U seasonally adjusted figure at each quarter end date.
Because of the way the time lag works, that means I’m actually using the CPI figure for May for the quarter ending June 30,
because that’s the last known-and-published figure available at June 30.
(May is also the middle month of the quarter, so that almost makes some sense? Reverse rationalization)
Convert all figures for that quarter to today’s money using the latest available CPI-U seasonally adjusted figure.
Look at the revenues or earnings or whatever by adding up any four consecutive quarters after the inflation adjustment.

Jim

4 Likes

The reason I adjust financial statements using CPI, is to compare apples-to-apples.
IMHO a dollar is worth what it will buy you at that moment. Not what BLS thinks the prices should have been.
Sure, energy and food are volatile, but can I go to a gas station and offer to pay the BLS-adjusted price? I cannot. So, to me it makes more sense to use the unadjusted dollars.
Maybe if you ALWAYS add four quarters and never compare quarter figures, then using adjusted dollars may be OK.

I meant unadjusted CPI, not dollars.

The reason I adjust financial statements using CPI, is to compare apples-to-apples.
IMHO a dollar is worth what it will buy you at that moment. Not what BLS thinks the prices should have been.

Many here do not like John Williams’ Shadowstats stuff, especially the stuff you have to pay for.
But his plots of inflation

http://www.shadowstats.com/alternate_data/inflation-charts

show both CPI-U (the official government figures) as well as his calculation (SGS, using the original 1990 or 1980 years as starting points. They are free.

So if you do not like his methods, ignore the blue lines on these two charts and look only at the red lines. If you do like his methods, look at both the red (CPI-U) and the blue (SGS -Shadow Government Statistics) values. These two charts start in either 1990 or 1980, depending on how you feel about things.

6 Likes