Monday, May 9, 2011

Setting Inflation Straight, Part III (at least)

You ever noticed how inflation seems a lot worse than the official numbers indicate?

Via Yahoo, Fox Business reports on the change in prices for a sample of everyday grocery items. It shows quite a shocking increase over the past year, far more than you might suspect from the "tame" inflation numbers you hear about.

I've reproduced the prices from that article in the table below, showing the current, March 2010, and March 2006 values. (I couldn't find the numbers in the source cited, but will operate on the assumption they all refer to March of that year, even though it suggests they average over 12 months in the previous year; this would mean the results I calculate actually understate inflation.)


Yikes! The average 1-year price increase for this sample is over 8%!

So what is the offical food price increase? The BLS CPI report on page 2 gives their aggregate 1-year food price incease as only 2.9%!!! And if you think 1 year is too short because of volatility, then look at the five-year food inflation numbers, a time period that covers the "massive" price collapse and "deflation" following the 2008 crisis onset: 4.7% per year.

And this is still:

- ignoring all quality debasements, and
- in an environment where banks are holding on to their massive reserves, suppressing price increases!

I've listed the corresponding prices for gold (though they're all relative to the present instead of March of any year), using an ETF (ticker symbol GLD) that tracks it. Looks like it works well (if a bit too well) as a barometer of dollar debasement. Hope you stocked up back then! (By a great coincidence, a financial advisor in April 2006 looked at me like I was insane for suggesting putting any money in gold.)

But don't worry, your iPad holding more memory will make up for this, I'm sure...

14 comments:

jsalvati said...

Let me start off by saying that I have a natural inclination to doubt this. However, your numbers did seem to support your case, so I mediated to myself 'if inflation is high I wish to believe inflation is high'.

A couple of points:

1) I for one have not had the impression that inflation is higher than ~2%, but I also tend to discount my perception on fine trends. Perhaps my powers of observation are not as good as other people's.

2) I'm not sure why you focus on quality debasements. The opposite holds for quality improvements. The direction of the overall effect is ambiguous. There are a couple of effects I can think of: 1) rising income -> quality improvements as people want higher quality foods (demand surface changes, seems to me obviously to lead to increased quality) 2) changes in technology make it more cost effective to produce foods of generally higher or lower quality (supply surface changes, seems ambiguous to me). I don't see a good reason to think that there's a downward bias.

3) This is a pretty small sample and it doesn't include weightings. It's evidence for inflation being higher than stated, but pretty weak evidence. The case for inflation being higher than stated inflation has to be made with much more systematic evidence and arguments. But you're in luck! People have done that! (http://www.shadowstats.com/, and I imagine there are others; there are also people who argue the opposite http://cafehayek.com/2006/02/does_the_cpi_ov.html). I haven't invested time figuring out what better inflation statistics would look like or what the problems with current statistics are, so I don't have a strong opinion. I would however be interested in a rationalist's investigation of such issues, however. This wasn't it, though.

3) I think gold prices is likely a totally uninformative statistics. Gold is very volatile; there's fairly large changes in demand (emerging markets), as well changes in supply.

I also redid (some of) your calculations:
https://spreadsheets.google.com/ccc?key=0Av25X_YdRE3TdFZtY1U1XzNGNGVHcWVSX2hWdk1BZEE&hl=en&authkey=CNTHxLUD

Dave Narby said...

Gold is hardly volatile compared to other commodities, or the stock market. Silver is volatile. Gold is not. Not sure why you would think this.

The CPI is gamed as it excludes 'volatile' things such as food, energy and housing. Which is naturally where the Fed caused inflation is showing up.

MIT's Billion Prices Project used to show many other nations actual price inflation compared to their CPIs, with few exceptions they all underestimated inflation. The BPP still shows US inflation, and no surprise, it's higher than the official numbers.

The US government has gone beyond jawboning a recovery, now they are attempting to fabricate one.

---

Check out the latest win for Liberty at http://libertyiswinning.blogspot.com/ !

Desolation Jones said...

You can actually find the source of their data here.

http://www.bls.gov/ro3/apmw.htm


I spent a few hours looking through the numbers and something I noticed is that food prices really are damn volatile. Prices could shoot up for one month, and then come down again the next month. They're even volatile year to year. For example, the average price of butter in 2010 is lower than in 2001!

So I decided to do some of my own calculations to see how the inflation rate using this fox business basket of goods varies year to year. One annoying thing is that the BLS doesn't provide annual data so I had to calculate it for each individual item myself in excell. There's only 3 months for 2011 so I left that out and only did 2001-2010.

http://i.imgur.com/BOAUn.jpg

Obvious both our results are not weighted so they're not very scientific, but it's interesting to look at nonetheless. The BLS numbers don't seem that different from mine to be honest. Over a 10 year period, the average yearly inflation rate is 3%. For 2010-2006, I got an average inflation rate for 3.71. It may not look that way because of the wild swings, but it does seem to average out to that lowish number.


“So what is the offical food price increase? The BLS CPI report on page 2 gives their aggregate 1-year food price incease as only 2.9%!!! And if you think 1 year is too short because of volatility, then look at the five-year food inflation numbers, a time period that covers the "massive" price collapse and "deflation" following the 2008 crisis onset: 4.7% per year. “

I have to say that I am surprised by the 2011 numbers. They do so seem a tad high. I believe the difference between the BLS 1 year inflation rate and your number is that they average the prices of the last 12 months and calculate the percent increase with the average 12 months prices of the previous year while you calculated the percent increase using the prices of a single month with the average of the entire previous year. If anything your method, overstates inflation because March could have simply been one of those months with exceptionally high prices. It wouldn't be surprising do to the the volatility I mentioned. And with commodity prices coming down, soon enough I'm guessing we'll have lower food prices.

Here's the annual data (google docs) if you want to play around with it. Hopefully my excel skills didn't fail me.


http://preview.tinyurl.com/4y53nmx

Desolation Jones said...

I posted a longish comment here yesterday. It seems to be missing now.

jsalvati said...

huh, that is indeed odd. I doubt Silas would remove it, so here it is (from my email):

You can actually find the source of their data here.

http://www.bls.gov/ro3/apmw.htm


I spent a few hours looking through the numbers and something I noticed is that food prices really are damn volatile. Prices could shoot up for one month, and then come down again the next month. They're even volatile year to year. For example, the average price of butter in 2010 is lower than in 2001!

So I decided to do some of my own calculations to see how the inflation rate using this fox business basket of goods varies year to year. One annoying thing is that the BLS doesn't provide annual data so I had to calculate it for each individual item myself in excell. There's only 3 months for 2011 so I left that out and only did 2001-2010.

http://i.imgur.com/BOAUn.jpg

Obvious both our results are not weighted so they're not very scientific, but it's interesting to look at nonetheless. The BLS numbers don't seem that different from mine to be honest. Over a 10 year period, the average yearly inflation rate is 3%. For 2010-2006, I got an average inflation rate for 3.71. It may not look that way because of the wild swings, but it does seem to average out to that lowish number.


“So what is the offical food price increase? The BLS CPI report on page 2 gives their aggregate 1-year food price incease as only 2.9%!!! And if you think 1 year is too short because of volatility, then look at the five-year food inflation numbers, a time period that covers the "massive" price collapse and "deflation" following the 2008 crisis onset: 4.7% per year. “

I have to say that I am surprised by the 2011 numbers. They do so seem a tad high. I believe the difference between the BLS 1 year inflation rate and your number is that they average the prices of the last 12 months and calculate the percent increase with the average 12 months prices of the previous year while you calculated the percent increase using the prices of a single month with the average of the entire previous year. If anything your method, overstates inflation because March could have simply been one of those months with exceptionally high prices. It wouldn't be surprising do to the the volatility I mentioned. And with commodity prices coming down, soon enough I'm guessing we'll have lower food prices.

Here's the annual data (google docs) if you want to play around with it. Hopefully my excel skills didn't fail me.


http://preview.tinyurl.com/4y53nmx

Desolation Jones said...

thanks!

Bob Roddis posted this in Bob Murphy's blog. Explains the missing comment.

http://www.webhostdir.com/news/ShowItem.aspx?ID=90246

Silas Barta said...

Sorry I haven't been replying to the comments here. (And w00t on the link from Bob.)

I just want to emphasize that I did not delete any post, and I do remember seeing the long one John reposted. I haven't even been on any screen where I'd get that option. Sorry it got deleted, I'll try to find out why. I'm glad it was at least saved.

While, I'm here, a few points:

@jsalvati:

1) You don't buy groceries, do you?

2) There is an asymmetry: the BLS adjusts downward for purported "improvements" but never upwards for debasements -- and they do not even maintin the labs (like Consumer Reports does) that would detect debasements. Stories about debasement have been all over the news. They *definitely* haven't factored in the plummeting of detergent quality due to the phosphate bans.

See the previous post linked at the end of my initial post for my explanation of the inconsistency with which they handle changes in quality: they've long told us that improved computers "aren't really deflation" (and therefore aren't an argument for a deflationary policy), yet are all two willing to import whatever computer hardware improvements they feel like in order to get the numbers to work out.

3) Yes, gold is volatile. Over long terms, it's still informative. Informativeness does not require perfection.

@Desolation: Just a few points:

a) By assuming the 2010 prices are focused in March rather than averaged over the whole year, I am *under*stating inflation, as I explained in the initial post. (And sure, March could be a nadir, but it's just as likely to be a zenith.)

b) Using a different weighting from the BLS has nothing to do with how scientific the results are; the weighting is a mostly subjective choice of emphasis.

c) All of my numbers err in the BLS's favor -- I'm just using the food prices they came up with, without incorporating the kind of adjustments and shortcomings documented by shadowstats ... my own data collection would probably make them look even worse.

***

And to all: even if you believe that your dollars are still buying all they used to, in the aggregate, you can still read my post as simply arguing, "Hey, here's why normal people _feel_ like prices are going up."

jsalvati said...

1) ha, you caught me.

2) Interesting, I did not know the BLS was asymmetric in their treatment of quality changes. Updated.

3) Right, hence I argued that it didn't tell you very much rather than nothing at all.

4) I want to note make it clear that the issue of whether the BLS under/overstates inflation is 90% orthogonal to our other macro disagreements. You could convince me of either one independently.

jfmappr said...

I don't know what kind of bread sells for $1.41, but I am reasonably certain I would NOT eat it... the bread I buy costs almost $4.oo.

Desolation Jones said...

"b) Using a different weighting from the BLS has nothing to do with how scientific the results are; the weighting is a mostly subjective choice of emphasis."

The BLS surveys thousands of people to see what type of items they buy the most to create their basket along with their weights based on that. It's not like a government bureaucrat simply subjectively chooses items based on his own preferences. At the very least you could admit it's more scientific than just arbitrarily choosing 10 items. I mean if you go by those 10 items only, we actually experienced 1% DEflation in 2005, which I would assume you think is hogwash.

I guess the difference between you and me is that I'm more interested in trends.





"I don't know what kind of bread sells for $1.41, but I am reasonably certain I would NOT eat it... the bread I buy costs almost $4.oo."

It's $1.41 per pound, not the whole loaf. If you buy fancier bread, then I'm guessing is wheat, which the BLS lists for $1.821 per pound.

John Forrester said...

I am just tired of using the worth;less,ineffective and malfunctioned mouses in my office. Employees are making mistakes,having problems in working and out output is getting worse day by day because of defective works. So.I think that I'll replace these one with best ergonomic mouse for office.

Payday Loans said...

On the off chance that you have no other choice than getting a payday advance, discover a place that has the best loan fees and strategies. A few organizations will remove cash from your paycheck to reimburse your advance in the event that you can't pay it back in time. Attempt to maintain a strategic distance from this if conceivable.

payday loans chicago said...

Also, you would prefer not to part ways with your friends and family. Thus, in the event that you acquire from a credit supplier, nobody from your friends and family will come to think about it.

car title loans online chicago said...

There is no surrender of a vehicle. It's an or more in addition to all around. An auto title credit is not as startling as it sounds if brought out with an authorized and mindful loan specialist.