Skimpflation: A reason inflation is worse than the government says it is

They present it as a way of looking at forex rates. If 1 USD buys you 1 swiss franc, but a Big Mac costs $5 here and 7 CHF there, then maybe the franc is overvalued. In any case you lose McDonald's purchasing power by converting.

But you could just as easily use it for inflation.

Big Macs are a stable product. A Big Mac sold in 1985 is very similar to one sold yesterday. Also, it includes a good mix of basic inputs. Land rents, transport costs, labor, and food ingredients.

Here's the raw data.

https://raw.githubusercontent.com/TheEconomist/big-mac-data/...

In the US, a Big Mac cost $2.54 in 2001 and $5.65 in 2021.

That works out to 4.0% annual inflation.

https://www.google.com/search?q=(5.65%2F2.54)%5E(1%2F20)&oq=...

The official numbers were 2% that entire time. Those come from a complex calculation using an always-shifting basket.

I'm not an economist, but one fundamental problem with any consumer index (as far as I can see) is that as you become richer in absolute terms the marginal utility of any additional dollar goes down (eg your life doesn't change that much if you upgrade your already expensive phone but it changes a heck of a lot if you can't afford to buy enough food) and therefore poorer people experience far more harm from inflation than richer people. Not sure any index ever captures that effect adequately.

Well:

> 2. How is the CPI market basket determined?

> The CPI market basket is developed from detailed expenditure information provided by families and individuals on what they actually bought. There is a time lag between the expenditure survey and its use in the CPI. For example, CPI data in 2020 and 2021 was based on data collected from the Consumer Expenditure Surveys for 2017 and 2018. In each of those years, about 24,000 consumers from around the country provided information each quarter on their spending habits in the interview survey. To collect information on frequently purchased items, such as food and personal care products, another 12,000 consumers in each of these years kept diaries listing everything they bought during a 2-week period.

> Over the 2 year period, then, expenditure information came from approximately 24,000 weekly diaries and 48,000 quarterly interviews used to determine the importance, or weight, of the item categories in the CPI index structure.

* https://www.bls.gov/cpi/questions-and-answers.htm#Question_2

The BLS' CPI isn't written on some set of secret scrolls. If you're curious about how it is configured just look it up.

The BLS isn’t hiding this information, it’s publicly available. The burden of proof would be on you to provide research on why that methodology is incorrect.

https://www.bls.gov/news.release/cpi.nr0.htm

> The CPIs are based on prices of food, clothing, shelter, fuels, transportation, doctors’ and dentists’ services, drugs, and other goods and services that people buy for day-to-day living. Prices are collected each month in 75 urban areas across the country from about 6,000 housing units and approximately 22,000 retail establishments (department stores, supermarkets, hospitals, filling stations, and other types of stores and service establishments). All taxes directly associated with the purchase and use of items are included in the index. Prices of fuels and a few other items are obtained every month in all 75 locations. Prices of most other commodities and services are collected every month in the three largest geographic areas and every other month in other areas. Prices of most goods and services are obtained by personal visits or telephone calls by the Bureau’s trained representatives.

The "basket of things" is made from the ratios of things consumers buy. BLS explains the methodology and lists the basket, and how it evolves as people buy less horse carriages and more cars, less giant radios and more iPhones, and so on.

Making a basket that better represents inflation would be worth a lot of money, and a lot of groups try, but none really do any better than the BLS.

For example, the Billion Prices Project out of MIT tracks a vastly larger number of things, but ends up with the same inflation rates http://www.thebillionpricesproject.com/

https://www.bls.gov/cpi/methods-overview.htm

Ever consider you are special? If you’re posting on this site, chances are you are more educated, richer, and better employed than most of your compatriots in whatever country you’re in. Of course luxury goods inflate faster. We can afford it.

And sometimes the content is actually superior. A US$ 30K nowadays is generally much better than a similarly priced car from 10 or 20 years ago. The iPhone was released in the United States in 2007 at the price of $499 for the 4 GB model and $599 for the 8 GB model: what kind of tech can you get for the same prices nowadays?

Everyone seems to ignore deflation, especially due to technology. This 1991 Radio Shack add illustrates the point quite well IMHO:

> There are 15 electronic gimzo type items on this page, being sold from America’s Technology Store. 13 of the 15 you now always have in your pocket.

> So here’s the list of what I’ve replaced with my iPhone.

> * All weather personal stereo, [*US*]$11.88. I now use my iPhone with an Otter Box.

> * AM/FM clock radio, $13.88. iPhone.

> * In-Ear Stereo Phones, $7.88. Came with iPhone.

> * Microthin calculator, $4.88. Swipe up on iPhone.

> * Tandy 1000 TL/3, $1599. I actually owned a Tandy 1000, and I used it for games and word processing. I now do most of both of those things on my phone.

> * VHS Camcorder, $799. iPhone.

> * Mobile Cellular Telephone, $199. Obvs.

> * Mobile CB, $49.95. Ad says “You’ll never drive ‘alone’ again!” iPhone.

> * 20-Memory Speed-Dial phone, $29.95.

> * Deluxe Portable CD Player, $159.95. 80 minutes of music, or 80 hours of music? iPhone.

> * 10-Channel Desktop Scanner, $99.55. I still have a scanner, but I have a scanner app, too. iPhone.

> * Easiest-to-Use Phone Answerer, $49.95. iPhone voicemail.

> * Handheld Cassette Tape Recorder, $29.95. I use the Voice Memo app almost daily.

> * BONUS REPLACEMENT: It’s not an item for sale, but at the bottom of the ad, you’re instructed to ‘check your phone book for the Radio Shack Store nearest you.’ Do you even know how to use a phone book?

> You’d have spent $3,054.82 in 1991 to buy all the stuff in this ad that you can now do with your phone.

* https://www.huffpost.com/entry/radio-shack-ad_b_4612973

That US$1600 Tandy 1600 runs a 286 CPU and has a 20MB hard drive, and supported 640×200×16 resolution (720×350 mode for monochrome monitors):

* https://en.wikipedia.org/wiki/Tandy_1000#Tandy_1000_SL_and_T...

What kind of system can you get for $1600 nowadays?

The basket can and does get adjusted.

I don’t care what a farrier charges to reshoe a horse or what a block of icebox ice costs, but I might have only 100 years ago. Surely we’d need a way to include entertainment and lifestyle changes as they happen. 1973 CPI didn’t include any costs for personal computers, Internet service, mobile phones, etc.

Also, consider that inflation numbers are necessarily lagging reality; they're not updated live by tracking residents or anything like that. For example, september's numbers are available and quoted above, but October's are not (and are projected to be higher).

This kind of conspiratorial thinking is harmful; it reduces trust, and thereby encourages everybody to skimp on the rules making everybody worse off. It's a classic prisoner's dilemma - we're better off just not playing that game. I'm sure the process of inflation-computation is imperfect, but let's talk about concrete, verifiable problems or better yet suggestions for improvements instead of nebulous feelings of disagreement.

By the looks of it, the official statistics confirm that inflation is spiking, but simply not yet to dramatic numbers (i.e. the kind of numbers that instantly change how we all think about money). If you feel that inflation is significantly higher, then why?

I'm searching that number and I can't find anything.

I found this, which is comparing the same quantities and has five pounds of meat. https://twitter.com/WhiteHouse/status/1410709115333234691

There's a bunch of complaints about other prices, and people pointing out the prices are up 8% since 2019, but I don't see anything about "cutting out all the meat" and I don't see the number 49 anywhere.

And the ground beef had the second biggest percent drop in price in that list, anyway.

Combine that with large grocery stores getting more data-driven it's kind of a no brainer for them to pick items they know they can increase the price of without blowback

And before you ask: No, it's not because we switched from buying prepackaged/processed food to raw ingredients. We just order-out less.

Some of it might have to do with the fact that I live in Chicago, a major logistics hub, in a middle class neighborhood that has price-sensitive consumers and multiple major grocery stores.

Oats, deli meat, fruit, pasta, that's about it. Canned beans have gone up a little, maybe butter too.

The biggest factor for us is changing what we buy, not prices. Buying less alcohol and fewer pre-packaged foods has made a huge difference in our grocery bill.

We haven't seen the claimed inflation here.

https://www.kroger.com/p/simple-truth-organic-grass-fed-beef...

Sure, now that travel restrictions are causing less uncertainty and demand is ramping up, this might not be repeatable. Perhaps hotels are skimping, not sure as we stayed with family. Either way, there's always some subjectivity in use cases and that can easily tilt the scale one way or the other.

But yes, the basket should be public information.

At least I can’t figure out how to game it meaningfully.

Many industries and businesses do this.

Adding housing maintenance, heating, water supply, etc. - the entire housing category climbs up to 17.8%. Which sounds like a totally reasonable estimate to me - young people and urbanites pay more, older folks and people in rural or provincial towns pay much less. From my personal experience - housing and related expenses were ~50% of my income while paying down mortgage, down to ~7% after the loan was repaid

Amusingly, highly educated people who are disdainful of anti-vaxers or Gwyneth Paltrow quackery can easily believe that inflation is an order of magnitude higher than officially reported, or that they the evil/incompetent bureaucrats leave out food, shelter, healthcare and/or energy costs from the calculation or some such.

As an academic economist turned machine learning/SWE, I'd say there are plenty of knowledgeable people around on HN, but on matters of economics and health, the signal-to-noise is definitely much lower than on anything related to computers.

1) PRIMARY factor. The rent is too damned high. It has been and people have been cramming together like sardines; but just like the car rush as PPE bubbles, now people want all the housing everywhere even MORE than the last 30-50 years when it wasn't being built sufficiently to accommodate demand by jobs.

2) The lowest end workers finally got enough of a lifeline, even if only for a short time, to realize how badly they were underpaid and how much they need to really have a modestly OK life. Combine with all the services that let them work in the first place (care for children, like daycare and schools) unable to provide due to the pandemic and this is hard-stalled between less workers and workers who hold out for more in what is finally a sellers market.

3) I don't have first hand observational data on this, but my gut feeling is that it's harder for businesses that used to do under the table work with citizens / undocumented immigrants to fully staff, and all the more likely for the pandemic to harm tightly packed workers in processing places. It would be nice to hear more about documented guest worker programs that legitimize migrant worker practices, while also ensuring that they're paid enough to discourage undercutting the local job market.

I'll also note: Rent affects the businesses too... and food prices might go up based on their (mistaken) belief that consumers can afford to pay more. I have cut back to lower quality options and rarer treats for some options in the last half year as prices have risen past the points I'm willing to buy at.

And as far as your rent going up by 30%, that's offset by some people whose rent actually went down by a lot during the early pandemic when people were fleeing the cities and bargains were to be had. And some people who don't pay rent at all, but pay fixed-rate mortgages which will go up by zero percent over the life of the loan. So you can see how that could significantly nerf the total inflation affecting that household.

Either it's a Swiss thing, or because we buy the majority of food semi-directly of organic farmers and prices haven't changed that much in this sector.

If your issue with the reporting of a statistic is that your personal anecdote disagrees with it, then you should probably look for more evidence.

If you open the link you shared, you can see they make hedonic adjustments for men's suits, the first item in the list. Are suits today really much higher quality than they were 10 years ago, necessitating a decrease in the actual price inflation of a man's suit? The fed thinks so, but they don't share the details of why or how much hedonic adjustment men's suits deserve.

First, they have to determine the conceptual basket of goods to track. Ok, that's comparable to your 1000 "highest volume" items methodology (albeit yours is a bit simplified too, we'll ignore the problem that good or volume is complicated for everyone). Generally stats bodies do this by looking at what consumers actually spend things on. This seems emminently sensible to me, and no offense intended, superior to a top 1000 volume method. Fuel, presumably, gets included in both.

Then they need to assign a weight for every good to determine how much its mixture of quantity + price movements contribute to the CPI figure. I put it to you this is necessary. Every item shouldn't be equally weighted even in your methodology (the distribution of volume of relative goods purcahsed/consumed is guaranteed to not be uniform across all 1000 goods, and it makes no sense for no. 1 to be given an equal weight to no. 999). Stats bodies do this too. This is what is meant by weight adjustment in this context. And both you and stat agencies should be doing it.

At some point, your "top 1000" goods is going to change and you're going to have to figure out what is in a new basket at a new time. So too does the CPI basket of goods. You both need to do this. Note that your methodology does not actually fix the fuel-budget-substitution issue. Price movements in your method that invoke a strong enough substitution or consumption effect result in fuel dropping out of your top 1000 as well.

By keeping your basket constant for a given period of time and just tracking the good's prices, you track expenditure somewhat consistently for a given period. That's a good thing. There's no reference to changes in a consumer's budget. But...CPI calculating stats agencies do this too for a set basket of goods just like you do. They fix the basket and track it over time, adjusting, like yours, only for re-inclusion the basket of goods under consideration.

Lastly, comes the hedonic quality problem. Changes in quality of consumption/goods over time. Some substitutions seem just because of taste or culture (see for example, relative consumption of antimacassars over time, or certain culture's preference for certain foods or items). Others are obviously qualitative in nature (i.e. computer processing power), and most things are a complex mix of tech/economy/culture/price/quality. This is the part that captures and tries to control for your qualitative substitution problem (it's not without its controversies, but I argue that your method needs it as well because it suffers similar problems, and indeed all ones which try to standardise subjective consumption of a national economy into a single figure). Relevant to 'fuel' is its price, but also substitution to other new fuels that turn up (hydrogen, LPG, lithium-ion), but also technology efficiency gains that have changed the consumption to utility ratio. Again, stats agencies try to control for this too.

Aside from the definition of the basket of goods (a weakness and methodology shared by your "fix"), there is generally no explicit reference to consumer's budget or assumed spending patterns when tracking within a given basket. There is nothing inherently equivalent to "assumed spare money or choosing to spend or not". (I suppose I'm happy to be proven wrong on this for a given countries methodology). it's not just a simplified survey of median households or expenditure.

Now, you can (and many people justifiably do) take issue with ideas and quantification of hedonic quality problems: but it's a universal problem, your method doesn't fix it, and I think you're generally misconstruing how CPIs are calculated around the world. The problems you mention are already attempted to be tackled and controlled for in methodology, and the official methodology is already generally superior to your suggested replacement.

Edit: it's also not like stats/economics agencies are doing this WITHOUT an accesible measures of nominal wages (that is wages and resource flows without inflation taken into account) available to them.

edit 2: in the real world, further considerations like geography, seasonality, population, etc is additionally controlled and adjusted for. And many stat agencies, in my experience, offer additional breakdowns of other price series, such as geographical and categorical indexes, or various additional weightings and combinations, so if you take issue with headline CPI you can investigate other more appropriate measures for your specific use cases.

What problem do you see in that approach? That looks exactly like the expected behavior from the average consumer.

If your goal is to track the prices of goods and services actually consumed by the population, isn't it appropriate to lower the impact of goods and services that are consumed less by said population?

> I've been down the rabbit-hole on CPI and learned a lot (...). The Wikipedia page is pretty thorough for US CPI [0].

Do you actually feel you're in any position to speak with authority on a subject, inclusive down to people with a background on economics, just because... You claim you read a Wikipedia article?

If they really want me to choose to save water they should offer a discount if I decline the service.