Comment by redwood | Hacker News Reader

Comment by redwood | Hacker News Reader

"at this point, all we can say with any confidence is that it does not appear that women have a higher perception of inflation than men because of the things they buy, the frequency of their shopping, or their knowledge of officially reported"

Really? This is ridiculous. Men and women do buy very markedly different things in general, especially in family life. For example, men disproportionally buy technological items which contribute significantly to the stable CPI.

CPI isn't just bread and butter... if it were inflation rates would be higher. Since the government can point to easier access to quality-of-life-enriching items like computers, which fall in price for equivalent goods year over year, it can manipulate the stats and make inflation look very low. This is appropriate for an "ideal" household that buys exactly that the CPI index holds.

But on the other and someone who buys disproportionally the items that are rising faster than the CPI really does feel a higher inflation rate.

It certainly seems conceivable that women tend to buy items that are rising in price faster than the CPI more than men.

It's also possible that marketing gimmicks disproportionally target women, with clothing coming to mind for example: perhaps sticker prices rise much faster than inflation, but post-"SALE" prices don't. This is a gimmick designed to encourage buying of deep sale prices, that may play better in clothing than in other industries, and that thus women fall victim to more. But since the sticker prices rise, they're noticeable.

Because CPI measures a rate, explanations like this only work in the short run. As computers/etc become cheaper, they become a smaller and smaller fraction of CPI. In the long run, CPI will be completely dominated by whatever factor grows the fastest.

The article provides some evidence that this is not a short run phenomenon, nor a phenomenon limited to the US. A similar gap is present among Swedes in 1977, and persisted in the US from 1987-2000.

As redwood alluded to, the content of the CPI is somewhat "gamed", though. It excludes "volatile" items such as food, housing and fuel. You know, the non-optional, inelastic purchases that people have to make.

No, CPI includes food, housing and fuel. There is a "core" measure that excludes those, since they add little information but a lot of noise when you want to track the underlying trend.

Core CPI includes housing, but not food or fuel (since they are volatile).

The difference between core CPI and exponentially smoothed CPI (which removes volatility but includes long term changes in food and fuel) is negligable:

http://www.newyorkfed.org/research/staff_reports/sr236.pdf

There may be gaming in CPI, but excluding volatile components is not it.

Another way to see it: CPI and Core CPI track each other pretty closely, and until recently Core CPI was actually higher than CPI: http://research.stlouisfed.org/fredgraph.png?g=cMs

It's because the CPI is actually gamed. You just have to change the basket of goods measured fast enough to keep the rate down. Like the substitution of hamburger for steak that Greenspan did. The website detailing how this "isn't shady" is laughable. http://www.bls.gov/cpi/cpiqa.htm

Check out http://www.shadowstats.com/: "One of my early clients was a large manufacturer of commercial airplanes, who had developed an econometric model for predicting revenue passenger miles. The level of revenue passenger miles was their primary sales forecasting tool, and the model was heavily dependent on the GNP (now GDP) as reported by the Department of Commerce. Suddenly, their model stopped working, and they asked me if I could fix it. I realized the GNP numbers were faulty, corrected them for my client (official reporting was similarly revised a couple of years later) and the model worked again, at least for a while, until GNP methodological changes eventually made the underlying data worthless. "

Steak -> burger -> dog food. Then what?

Most of the "tricks" proposed by conspiracy theorists for tweaking CPI only work once or twice.

Right, obviously you can only go so far on the steak to hamburger. But there are lots of places for abuse. The problem is that the definition for the CPI keeps changing, and thus, it's meaning over time is less useful.

What this study showed is that perceived inflation is real and doesn't jive with the numbers that the economists produce. Who is right? A 500 person sample will be good to predict the sentiment of the US as a whole. Gallup says that they only need 1,000 people to get +-4% so this is reasonable.

Ask yourself, which is more likely? The entirety of the US population doesn't understand how prices affect them day-to-day. Or that some economists have devised a moderately useful but not flawless metric?

Price formation comes from the relative values that every human being places on the marginal unit of a good. Prices are an emergent property of people interacting chaotically in a marketplace. To suggest that literally EVERYONE's ability to sense the change in price over time you also need to throw out the very people that set the prices to begin with.

EDIT: Link to gallup http://www.gallup.com/poll/101872/how-does-gallup-polling-wo...

Given that the US population can't even agree with itself, it's likely that the US population does not, in fact, understand how prices affect them day to day.

Or, more precisely, the US population does not accurately remember prices from years ago. I definitely couldn't tell you how much milk cost in 2005.

Right but these people are getting polled regularly, not every five years. So they don't have to remember how much milk has gone up since 2005, but rather how much milk has gone up in the last 3-6mo. It's not easy but it's not impossible.

Isn't it more about shifting what constitutes the CPI? In other words, before computes, 0% of the CPI was from computers. However now, every year going forward, computers become more and more important as a constituent of the CPI. On the flip side, computers tend to get cheaper. Don't you see potential for creative CPI engineering here? Now don't get be wrong: computers are like food today--quite fundamental. So I understand why they do it. I'm just pointing out that this can lead to the CPI being out of sync with things like food prices. No big deal, it can be so.

Do you actually believe their numbers? If they where true, it would mean that the US has been in a depression for well over 10 years. That would be the longest depression in the history of the US, but somehow nobody noticed because the BLS managed to cover it up. It would also mean that there was never any housing boom, house prices where stable relative to other prices.

Do you actually believe this?

I don't know where you went from "here look at this link, it's interesting" to "we've been in a recession for 10 years" but I would tend to agree with you actually.

It's not that the BLS covered anything up, but rather than the CPI is a flawed metric and it's continual revisions always go in the favor of the central bank.

When the dot com bubble burst we had the first ever recession where housing starts and car sales didn't decline. Doesn't that seem strange to you? That smells fishy to me.

I'm not saying that the BLS are a bunch of evil geniuses who through brilliance and skull-duggery keep us all in the dark. I think it's much more plausible that a few key people jigger things to do political favors to others and it makes the CPI look lower than it should.

It's the interaction of everyone in a marketplace that sets prices to begin with. Without people, you can't have prices that are meaningful. So if you want to say that people aren't competent to understand inflation, then it seems to me that you also have to say that people aren't competent to set prices. Since the latter is not true (by definition), I'd suggest that the former doesn't hold either.

Regarding the bls.gov link you linked....

1) The link explicitly claims that hamburger was not substituted for steak (and the shadowstats link does not contradict this, only claims that it was proposed). So why are you claiming that Greenspan did perform this substitution?

2) I don't find their explanation laughable at all; I find it entirely plausible. And, without having read all of the shadowstats site (only about half of the article on the CPI), I found it ... long on wild claims, short on methodology or references.

With this in mind, I'd be curious to hear a clarification of what you meant, and why you believe it.

A regular average would suggest that the basket of goods used to compare from interval to interval does not change. In other words, a 1lb ribeye in 1950 is the same as a 1lb ribeye today (in quantity, not price). If it's included in the CPI then the price increase is reflected properly.

Using a geometric average implies that the basket of goods selected CHANGES depending on the change in prices. So in 1950 when a ribeye was cheap, 1lb of it was included in the basket of goods. Now that it's more expensive, only 0.5lbs of it are included in the basket.

This is a made up example. Prices are not correct. Just for illustration. 1950: hamburger - $0.50/lb ribeye - $2/lb arithmetic mean: (2 + 0.5) / 2 = $1.25 geometric mean: sqrt(20.5) = $1.00

2012: hamburger: $3/lb ribeye: $18/lb arithmetic mean: (3 + 18)/2 = $10.50 geometric mean: sqrt(318) = $7.34

Arithmetic inflation: 10.50/1.25 = 8.4 Geometric inflation: 7.34/1 = 7.3

The geometric mean has two wonderful properties: 1. It makes the average price lower by adjusting the quantities included based on their relative prices 2. It makes the inflation over time seem lower as well

I'm not saying that this is some giant conspiracy by "the man" to "keep us down" but I would argue that it's intellectually dishonest.

In order for the CPI to be useful to make comparisons from one year to another the quantities of goods in the basket have to remain constant. Since the geometric mean effectively varies the quantity in the basket (relative to arithmetic) I would argue that it robs the CPI of any power to provide real comparisons.

You are correct that for chained CPI to measure inflation, the basket of goods would need to remain constant. Among other things, if we want CPI to be well defined in 1970, then CPI in 2012 would need to exclude all goods invented after 1970.

If we did this correctly, inflation would likely be drastically lower. The biggest driver of inflation is health care, and 1970's health care is pretty cheap.

In fact, CPI only measures instantaneous inflation - the inflation rate of today's basket of goods. It is fundamentally incorrect to treat chained CPI as inflation, particularly over long periods.

If you actually want to measure inflation between two endpoints, you'll actually need a formula which depends on the starting year - i.e., you'll have CPI1970 which will be fundamentally different from CPI1990 or CPI2012.

If a bunch of physicists came out with a paper entitled "The Curious Case of Every Electron Having Slightly the Wrong Charge" they'd be laughed out or arxiv.

How come some economists can say that the system they are studying is "wrong" relative to their models and be taken seriously? If your model doesn't line up with the data, it's wrong!

Reply on news.ycombinator.com