10 ms·
Some people would naturally respond to the content of the article. Others would naturally appeal to authority or ad hominem the author. Some food for thought.
by lend000 5y ago
Some people would naturally respond to the content of the article. Others would naturally appeal to authority or ad hominem the author. Some food for thought.
CPI ostensibly worked the same way in 2004, by the way. Based on your other comment linking the definition of CPI, it seems like you have no idea how it's calculated. Even I didn't know the details until recently -- I just knew the numbers weren't matching up with my models and very obvious price data.
So here's how it works. In the US, a consumer expenditure survey is obtained by the BLS [0]. This determines what items account for typical Americans' spending, and in what quantity, to determine weights [1]. Price data is obtained, with in-house selected indices and hedonic adjustments [2]. These prices are multiplied against the weights, summed, and reflected as a ratio compared to the previous year. The Wikipedia page provides a nice summary [3].
What is the problem with this? Well, the problem is in the consumer expenditure survey.
If wealth inequality is stagnant, this is actually a decent measure of inflation in so far as it affects people's personal expenses. However, if wealth inequality is increasing (and the assets and expenses of the wealthy are not factored into this survey, which they are not), inflation can only increase so much as median Americans' wages increase (plus their debt). When prices are increasing, Americans buy dramatically different amounts of the expensive items (you can see that in the weights in [1] in the mid 2000's when gas prices varied significantly) to fit their budget, under-weighting the items that are manifesting high inflation. Inaccurate hedonic adjustments and any other adjustments done behind closed doors could also account for why CPI veers so far from independently calculated inflation measures like the Big Mac Index, but the core of the problem is that the inflation weights are reactionary to inflation and there isn't much of a delay on letting weights change each year.
So, that is why real wages adjusted for "inflation" is a straight line.
[0] https://www.bls.gov/cex/ https://www.bls.gov/cex/
[1] https://www.bls.gov/cpi/tables/relative-importance/2019.pdf https://www.bls.gov/cpi/tables/relative-importance/2019.pdf
[2] https://www.bls.gov/cpi/quality-adjustment/questions-and-answers.htm\ https://www.bls.gov/cpi/quality-adjustment/questions-and-ans...
[3] https://en.wikipedia.org/wiki/United_States_Consumer_Price_Index https://en.wikipedia.org/wiki/United_States_Consumer_Price_I...
- arcticbull 5y ago> Some people would naturally respond to the content of the article. Others would naturally appeal to authority or ad hominem the author. Some food for thought. This was a general comment on the quality of Forbes opinion pieces not an ad hominem attack. It's worth mentioning it because I think there's a strong tendency to trust the Forbes brand, but it doesn't extend to their blogging platform. That's why I actually separated it from my other reply. Inflation is defined relative to basic goods and necessities, not to all goods - or to luxury goods, and there's a reason for that. Prices of goods change all the time for all sorts of reasons. Lobster used to be garbage food, now it's luxury. Same with sushi, monkfish, oyster and pig trotters. If the quantity of lobsters in the ocean start going down and the price of lobster increases, does that mean inflation is happening? Even if people switched from lobster to salmon? Certainly not in my opinion. Nobody's not entitled to a constant basket of necessities isolated from the world around them. Basing the inflation definition on basic goods and needs at the time is a reasonable proxy due to their broad-based requirement by individuals. Nobody needs luxuries by definition. Luxuries are expensive by definition. Luxuries tend to be scarce and their price as a result fluctuates substantially. But further, they're a product of social tastes that change for any reason or no reason. Weights are adjusted because the price of things changes for all sorts of reasons, and the goal is to index against a basket of common necessities - not specific necessities for that reason. A good analog would be the S&P 500 and the Dow Jones. Those index components change regularly as some companies fall out of favor and others rise to prominence. They reflect the economy but not specific companies. That's not a bug, it's a feature IMO. The weights being adjusted is reactionary to the cost and availability of goods and the preferences of consumers, and that's totally fine. A platonic, isolated increase in price is not inflation, it's just what it says on the tin - an increase in price.
- reedjosh 5y ago> Basing the inflation definition on basic goods and needs at the time is a reasonable proxy due to their broad-based requirement by individuals But what of non-luxury goods that aren't included like housing? If the CPI is to measure essentially the value of the dollar to what is required to live, isn't it useless without housing? Why is housing left out? I surmise it wouldn't fit the inflation targets the fed claims to be attempting. Either way, when a measure becomes a target it fails to be a measure. Edit: Just another thought. Is today's produce even equivalent to that of 60 years ago? Today's corn is genetically modified and soaked in `safe` glyphosate. If the CPI was honest they'd at least have to acknowledge that the food we eat today is of extremely so lesser quality. Today's nearest equivalent would be to eat a wholly organic diet. If one were to hold the quality of food stable, that too would have to show a _much_ higher rate of inflation.