Current inflation doesn't ..

[pinqy]

That's exactly what he does.
He doesn't collect any data of his own.
He does collect plenty of money from his idiot subscribers.

My favorite is looking at his Unemployment rate, because we can do the math to figure out the number he's adding. To get from a U-6 of 14.9% to his 22.4%, you have to add 15,132,000 to the numerator and the denominator. And yet the number of people not in the labor force who say they want a job now is only 6,378,000 and 2.6 million of them are already in the U-6 as Marginally Attached. So that's 11.4 million more people than even want a job but he claims he's only adding people Discouraged more than a year.

 

[iamwhatiseem]

editech is dead on.
The CPI is more or less a useless figure when trying to gauge consumer experience.
It is too broad in both data measurements and demographics.
America is too big and far, far too diverse to come up with a meaningful average.
There is no such thing as "an average American".

 

[pinqy]

editech is dead on.
The CPI is more or less a useless figure when trying to gauge consumer experience.
It is too broad in both data measurements and demographics.
America is too big and far, far too diverse to come up with a meaningful average.
There is no such thing as "an average American".

Good thing it's not supposed to gauge consumer experience then. It's supposed to measure the overall price changes for the nation for a general direction. And it is good at that.

 

[iamwhatiseem]

editech is dead on.
The CPI is more or less a useless figure when trying to gauge consumer experience.
It is too broad in both data measurements and demographics.
America is too big and far, far too diverse to come up with a meaningful average.
There is no such thing as "an average American".

Good thing it's not supposed to gauge consumer experience then. It's supposed to measure the overall price changes for the nation for a general direction. And it is good at that.

Until you try and define "overall price changes".

 

[pinqy]

editech is dead on.
The CPI is more or less a useless figure when trying to gauge consumer experience.
It is too broad in both data measurements and demographics.
America is too big and far, far too diverse to come up with a meaningful average.
There is no such thing as "an average American".

Good thing it's not supposed to gauge consumer experience then. It's supposed to measure the overall price changes for the nation for a general direction. And it is good at that.

Until you try and define "overall price changes".

Touché. That's obviously difficult and controversial. Certainly, you have to use an index. But which one? That depends a lot on time frame and other collection. Some indexes use current weights, which means you can't be as timely. There's a LOT of research and argument for best index.

 

[editec]

Price changes are meaningless if you don't ALSO understand what percentage of the family income gets spent on the different items in the CPI basket.

A 10% increase in the price of energy effects a family in the bottom quintile of income FAR DIFFERENTLY than it does families in the 2nd, third, forth and fifth quintiles. And as their basket of good is so wildly different, so too wil the CPI computed for those different quintiles be wildly different

And no additional data would be necessary to CAPTURE the different CPI outcomes if the models reflect the very different BASKETS OF GOODS that each quintile typically purchases.

Seriously, kids, what I am proposing is neither expensive, confusing or remotely beyond the ability of the BLS to compute.

Let me propose that the entire reason this is NOT done is because America HATES to admit that there even are socioeconomic CLASSES in this nation.

Of course that is absurd on its face, but BIG LIES like the classlessness of America are cherished lies that we like to tell ourselves.

 

[pinqy]

Price changes are meaningless if you don't ALSO understand what percentage of the family income gets spent on the different items in the CPI basket.

You can get a lot of that from the Consumer Expenditure Survey

A 10% increase in the price of energy effects a family in the bottom quintile of income FAR DIFFERENTLY than it does families in the 2nd, third, forth and fifth quintiles. And as their basket of good is so wildly different, so too wil the CPI computed for those different quintiles be wildly different

Exactly...which is why it would be expensive and difficult to do.

And no additional data would be necessary to CAPTURE the different CPI outcomes if the models reflect the very different BASKETS OF GOODS that each quintile typically purchases.

You would need additional data to get the weights and expenditures for a price index. While the CE gives expenditures, that's not directly translatable into actual indexes.

Seriously, kids, what I am proposing is neither expensive, confusing or remotely beyond the ability of the BLS to compute.

It is expensive and would require a lot more work.

Example: Read the following article on The experimental consumer price index for elderly Americans (CPI-E): 1982–2007] Summing up the methodological issues:

1. Because the sample size of elderly is small, the weights have higher sampling error.
2. The same outlets are used as for the CPI-U which does not accurately reflect buying patterns of the elderly.
3. The items selected are proportional to probability for the overall population and don't accurately reflect buying patterns of the elderly.
4. Senior discount prices etc are only used in proportion to the entire population which doesn't reflect elderly prices.

For any sub-sample of population you'll run into the same problems of higher sampling error, poor reflection of outlets used, and poor reflection of buying patterns.

To fix any of those problems, you'd have to expand the Consumer Expenditure Survey, expand the Telephone Point of Purchase Survey, expand the CPI collection, AND hire more people for all three surveys to handle the larger amount of data.

You keep saying BLS already has the data, but they don't.