Hacker News Reader

Comment by missedthecue | Hacker News Reader

I agree in general about the hedonic adjustments and their potential for manipulation or miscalculation. But automobiles are probably the worst example!

Toyota didn't even roll out passenger side airbags until 1998. Only the driver had them prior to this. The new Corolla has 8 airbags. Today we have better fuel efficiency, collision protection, driver assistance, remote start... Even the seats are more comfortable. Not to mention most car brands have much better warranties today than they did in the '90s.

Other things get hedonic adjustments but shouldn't. Shirts for instance.

View on HN

But it's irrelevant when that's the cheapest new vehicle, most working Americans need a car, and despite better fuel efficiency, it isn't really that much cheaper to own a car. Hedonic adjustments make sense in theory, but in practice it's a loophole to make inflation whatever the government wants it to be. In the case of vehicles, they clearly went overboard. The counterargument to hedonic adjustments is that almost everything should become cheaper over time, including all vehicle parts and assembly, due to technological progress in manufacturing. I don't think it's unreasonable to assume hedonic adjustments should cancel out increases in manufacturing technology and just ignore them both, probably resulting in a simpler AND more accurate CPI.

> But it's irrelevant when that's the cheapest new vehicle, most working Americans need a car, and despite better fuel efficiency, it isn't really that much cheaper to own a car. Hedonic adjustments make sense in theory, bpractice it's a loophole to make inflation whatever the government wants it to be.

The average new car in the US lasts like 2x longer than it did in 1998. Shouldn’t that factor in?

“The average new car in the US lasts like 2x longer than it did in 1998. Shouldn’t that factor in?”

Absolutely not. Where do you stop with this game? Do you factor in that the seats are better so people go less to the doctor for back pain?

Yes, that sounds like a good idea if what you're trying to estimate is the cost of maintaining a certain quality of life.

If you push this far enough then you can argue inflation away from almost anything.

So what? The goal of measuring inflation isn't to compute the biggest, scariest number, it's to track changes over time.

“I pay $30k for something I have to buy once every 10 years and that is ‘inflation’ vs. paying $20k for something I have to buy every 5 years”

I mean, I really struggle to follow your logic.

If a paper towel roll has 2x the sheets but costs 1.5x more, is that inflation too?

A lot of people buy a car every n years. For them it doesn’t matter if it lasts longer. How do you quantify how much longer cars last? Maybe people use them longer because cars are better or maybe they use them longer because they can’t afford new cars.

The paper towel is a bad example because you clearly can quantify the number of sheets. It gets more tricky if you say new paper towels are more absorbent so you need less so there is no inflation although the sheet costs more. How do you measure that? These metrics are way too easy to be gamed. Next is apples costing twice as much but they are crunchier so no inflation?

> A lot of people buy a car every n years. For them it doesn’t matter if it lasts longer.

You can’t just keep randomly changing the relevant populations. First it was working people who can’t afford a breakdown. Now it’s wealthy people who choose to buy cars regularly. Maybe that’s why it’s difficult to grasp inflation: it’s a basket of everybody

> How do you quantify how much longer cars last?

https://ihsmarkit.com/research-analysis/average-age-of-vehic...

> Maybe people use them longer because cars are better or maybe they use them longer because they can’t afford new cars.

You’re implying that people used to junk old cars because new ones were so cheap? Interesting take.

> The paper towel is a bad example because you clearly can quantify the number of sheets

I can also literally quantify how much longer cars last (see link above).

> It gets more tricky if you say new paper towels are more absorbent so you need less so there is no inflation although the sheet costs more.

Yeah, you know inflation takes a basket of spend into account, right? So yeah if in a year, the average consumer buys half as many paper towels that cost 2x as much that is exactly “no inflation”.

> These metrics are way too easy to be gamed. Next is apples costing twice as much but they are crunchier so no inflation?

I really suggest you read up more on the thinking behind inflation if you’re going to have such a strong opinion about it. Apples being crunchier is not accounted for in hedonic adjustments [0]. Like literally. Go to FK011

[0] https://www.bls.gov/cpi/quality-adjustment/home.htm

EDIT: Learn something new I guess. Cars aren’t even subject to hedonic adjustment. They have their own adjustment system that seems pretty fair if you read it[1]

[1] https://www.bls.gov/cpi/quality-adjustment/new-vehicles.pdf

I think the original poster's claim of lasting twice as long is a big exaggeration, but it'd be interesting to see a source. Regardless, cars are likely more expensive to maintain now with all of the added complexity / sensors that can break (remember when everyone could work on their own car? I bet they didn't factor that negative into hedonic adjustments.)

Surprisingly only a little!

The USDOT tracks this! That’s why cars are such a bad example.

In 1998, the average age of a car on the road was 8.8 years. [0]

In 2022 it’s 13 years. [1]

Factor in install-base effects and the multiple probably rounds to 2x

[0] https://www.bts.gov/content/average-age-automobiles-and-truc...

[1] https://www.carscoops.com/wp-content/uploads/2022/05/SP-Glob...

That's actually a pretty big difference compared to the OP's claim, AND note that this statistic incorporates a delay (we are comparing 1990 vehicles to 2009 vehicles instead of 1998 to 2020). Granted, we don't know how long 2020 vehicles will last yet. The final caveat being the second point I made about people holding onto their cars longer as inflation rises / changing their preferences, not necessarily because the cars are holding up better.

> The final caveat being the second point I made about people holding onto their cars longer as inflation rises / changing their preferences, not necessarily because the cars are holding up better.

I’m sorry but the idea that “we used to junk cars sooner because new cars were cheaper” is a very big stretch to support your bias (much moreso than the concept of hedonic adjustment)

> we used to junk cars sooner because new cars were cheaper

This is not obvious to you? Look at how many things (especially plastics and electronics) are wasted today while they still function because they are so cheap. How many times have you replaced your smartphone while it still worked fine?

The point, again, is that we can do reasonable mental gymnastics to more or less counter-act any hedonic adjustments (which were also mental gymnastics to make inflation seem lower in the first place). A better metric would involve fewer loopholes like this that allow arbitrary mental gymnastics in the first place and instead say "how much does an entry level vehicle cost, amortized with upkeep costs and fuel for the average American." There should be no adjustment for better airbags or bluetooth, nor a reverse adjustment for taking into account improved automation in factories.

> This is not obvious to you?

It is very much not obvious to me because it’s not true. Critical safety features like airbags (required by US law in 1999) and ABS (required by US law in 2000) were not standard in cars back then, which is a huge incentive to junk a car.

Since then, mandatory safety features include traction control (2011) and backup cameras (2018), and idk about you, but I’m fine driving a car without Either of those things, but not ABS or Airbags.

Honestly, if you’re going to have such strong opinions about cars, maybe do research into them.

And to your amortization point, literally amortization means “cost over useful life” so we agree?

And please look at the link on my other comment [0] because that is not how car prices are adjusted by the BLS

[0] https://www.bls.gov/cpi/quality-adjustment/new-vehicles.pdf

Do you think most people wouldn't get new cars more often if they were free?

It's possible to have many correct points but miss the forest for the trees/argue past each other. I didn't mean that last comment to be condescending, by the way.

To my original post (the second part), the BLS primarily measures the spending of people who never save a substantial portion of their income. Vehicles are a necessity for most of these people, so they are going to stretch out the lifetimes of their vehicles if the prices increase relative to their wages. Furthermore, people who might have bought a Camry with some upgrades in 1998 might be relegated to a stock Corolla in 2020. Sure, things are getting nicer all the time, but it's getting easier to make things all the time, too. Technology goes both ways, and at the end of the day, it's still a car. If they took out the rear airbags and removed all of the cheap microcontrollers, would the price really drop 40%?

The way the BLS does its survey does not account for preference changes due to price increases. When wealth inequality is constant, it's probably a decent measure. But even then, it's a political metric and the Big Mac index is the best, simple metric we have for true inflation (which has been consistently higher than CPI for decades).

If anything, shirts should get negative hedonic adjustments as the quality of the fabrics and craftmanship decrease. But I’m guessing negative hedonic adjustments isn’t a thing.

Yeah, like many other things. Tomatoes of today do not smell and have no taste. Nobody would buy such tomatoes 40 years ago even for half the price.

Reply on news.ycombinator.com