Hacker News Reader

Comment by caesil | Hacker News Reader

I don't really understand the argument at the heart of this article, which is "we should include interest rates in CPI".

How do interest rates effect everyday people exactly, other than price inflation on goods and services (which is included separately in CPI)?

The only way seems to be interest rates on personal loans and mortgages. So if anything, we should only include interest rates in proportion to how many people are taking out major loans during the sampled period (and maybe some additional amount based on the effect on adjustable-rate mortgages, etc).

Blindly stacking interest rates on top of CPI doesn't really make sense as a measure of personal inflation, and "it feels like stuff got more expensive" (as a lot of other comments here argue) isn't so much an argument for this strategy so much as an argument that the CPI 'basket of goods' needs to be rebalanced in other ways.

View on HN

> "The only way seems to be interest rates on personal loans and mortgages."

Yes, exactly.

As noted in the article and the underlying study, 80% of car purchases are done via a loan and financing is not part of the inflation measure. When you look at buying a home, most are also done via loans and still financing is not part of the measure. In both cases, interest rates are a factor in affordability and cost.

> Blindly stacking interest rates on top of CPI doesn't really make sense

Well then it's good that's not what they're doing.

I think this is valid for car purchases.

I don't buy this is reasonable for home purchases. They already factor imputed rent in CPI; you'd have to somehow do some sort of complex weighing of own vs rent to factor mortgages (complex since it only affects new purchases given most folks are on fixed rate mortgages)

> interest rates are a factor in affordability and cost

Paradoxically over a long term, interest rates don't affect house affordability much.

People bid on houses at the limit of what they can spend - the constraint is their income not the interest rate.

As interest rates fall, people pay the same interest payments but bid higher on the house price (driving house prices up). As interest rates rise, people spend the same amount monthly (on interest payments) but borrow less in total and can bid less on houses.

It is a steady-state argument, so other things do matter (income changes, mortgage qualification rules, immigration into the area, dynamic effects of interest rate changes). Rent has other factors but the constraint of income has parallel effects.

That doesn't seem to be true now. Home prices have risen despite higher interest rates.

That is irrelevant to what I wrote.

If interest rates remain the same for a longer period then houses must sell for a price where people can pay their mortgages.

Household income is a hard constraint. People can't pay more for interest payments than they earn.

House prices can still rise if:

• people can pay more on interest e.g. reduce their spending in other areas e.g. increase income using overtime. e.g. rent or AirBnB rooms.

• Or if they can lower mortgage interest costs by increasing deposits/equity e.g. sell down other investments, or borrow from friends/family e.g. downsize house.

• Or if more wealthy people move into the area

And it is a market - I am talking about how individuals drive the market price but obviously the price is set by successful sales and purchases. There is a lot of unmet demand by people who can't quite afford a deposit/mortgage.

> If interest rates remain the same for a longer period then houses must sell for a price where people can pay their mortgages.

No, houses can stay expensive even if no one lives in them, as we see in China.

Or rents can rise far above affordable levels, as they have, and people can end up paying more in rent than they could qualify for with a mortgage.

Or people can take roommates or boarders to afford the higher costs.

Irrelevant whether someone is living in it or not.

And I don't think China is a relevant example when discussing mortgages in my country or the US.

If the market is driven by mortgages (presuming there are not other reasons why the market is disfunctional) then someone is paying the mortgage (or the opportunity cost of investing cash).

The market price mostly depends on how much people are capable (and willing) to spend on interest payments. Somewhat different dynamics in New Zealand because to an approximation our mortgages are all variable so we see house prices shift over a couple of years as interest rate changes (plus our interest rates are often higher than US rates). US 30 year mortgages at a fixed rate mean market prices drop more slowly, but I'm guessing can still rise quickly. Interest rates went up in NZ and two years later property prices are down 10% currently I think (plus sentiment still really matters too).

You can still fiddle with terms and a bunch of other free variables, but interest payments are one of the most significant variables.

We had an earthquake in my city and uninsurable houses could not get mortgages and the price to buy one of those houses dropped drastically. https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...

> Or people can take roommates or boarders to afford the higher costs.

I agree with you so much that I already previously mentioned that ("rent or AirBnB rooms") in the comment you are replying to!

You did mention it later but not in the original comment I said wasn't true now:

> As interest rates rise, people spend the same amount monthly (on interest payments) but borrow less in total and can bid less on houses.

Your later responses only explain (in part) why that's not true.

The argument is bunk, and your observation is correct: inflation can be higher than the current CPI predicts, but this does not somehow imply that the CPI basket should factor instruments that do not disproportionately affect ordinary Americans' finances (or double-count ones already accounted for more directly).

> So if anything, we should only include interest rates in proportion to how many people are taking out major loans during the sampled period

Is that right? What about people that didn't buy a home in 2023 because interest rates were high (i.e. they the impact of high interest rates was so large that they _wouldn't appear_ in your weighting because they were pushed out of the market)?

Mortgages aren't some small detail distorting the numbers. They make the difference between being able to afford a house or not. Having to live with your parents vs being able to buy a house and start a family of your own.

I was lucky to buy a house 10 years ago, and got a 3.5% fixed rate mortgage. Today the house value has gone up by 50%, and mortgage rates are about double at 7%. Just due to the mortgage increase alone, someone buying this house today would have a monthly payment about double what I am paying. I would not be able to afford it. On top of the mortgage rate increase, there's also that 50% inflation in the value of the house, meaning that if it sold today the new buyer (assuming they financed it) would be paying closer to triple (150% x 200%) what I am.

Including interest rates in CPI muddles the distinction between the price of something vs how to fund the acquisition of something.

Interest expense is a real cost people pay. Why ignore it?

The reason to ignore it would be mechanical: it's more difficult to measure something when the measurement includes a thing that is adjusted by the measurement.

For purposes of setting Fed rates, it makes sense to exclude it.

For purposes of measuring perceived inflation, it makes sense to include it.

"inflation rate" is a nebulous thing.

Interest expense and interest rates, concretely paid by people like any other expense is not any more nebulous than tracking other expenditures.

We already know the interest rate and various categories of interest expense paid by population. It is just not included. Your rationale sounds very much like "we shouldn't include it because the inflation rate would be higher".

There is nothing mechanically weird about not ignoring a real expense the economy and people bear the cost of.

This isn't touchy feely, this is concrete costs people pay. Not tracking is purely political.

> Your rationale sounds very much like "we shouldn't include it because the inflation rate would be higher".

What is the CPI actually used for?

To set various levers at the central bank level to manage inflation and keep the economy humming along.

One of those levers is... the Fed funds rate... which in turn influences most other interest rates.

So you'd be incorporating interest into the measurement used to set interest.

> What is the CPI actually used for?

Technically, they pay closer attention to PCE.

CPI is used for other things too, including contract negotiations and standard of living comparisons.

Anyway, recursive sums like that are nothing unusual; as long as the sum converges it's not a problem. Any negative feedback amplifier does the same thing.

If in a year a burger goes from $5 to $10 and the Fed rate goes from 2% to 5%, what do you think the inflationary factor should be in your methodology?

It isn't exactly. Increasing interest can mean you have to pay liabilities unexpectedly early. But it's really the difference between your income and interest that matters. If your income is keeping up with inflation then higher inflation doesn't make you worse off; it just means you are paying off debts faster.

> The only way seems to be interest rates on personal loans and mortgages.

Most interest rates across the economy are set in relation to a benchmark rate, which loosely follow the Fed rate. E.g. the US prime rate.

This effects essentially all credit that isn't fixed rate, which is a huge portion.

How this effects you -- the price of credit that businesses use to function, which essentially every business uses, ultimately shows up in the cost of goods.

> So if anything, we should only include interest rates in proportion to how many people are taking out major loans during the sampled period

It goes both ways: in that case, we should also discount from CPI calculation the effect of sub-3% mortgages, low/zero interest auto loans, forgiven fraudulent PPP loans, etc.

Interest rates impact the cost of everything you buy.

Almost all large businesses are financing their operations on credit, not by spending down a war chest replenished with revenue. Large public companies borrow money against their remaining held stock to finance their operation. It is true that inflation impacts the base cost of the raw materials and labor but those costs are also more expensive because of the higher business loan interest rate to finance an operation. The extra financing cost is passed to the consumer.

The higher cost of financing drives layoffs too, companies will layoff when financing costs rise so they can stay cost neutral.

To me it makes sense to include interest rates, because even if a person doesn't have any loans (or get new ones), they pay current rates on the cost of loans, because businesses do have loans and they pass the costs through.

> we should only include interest rates in proportion to how many people are taking out major loans during the sampled period

The price of beef went up. A lot. As a result, quite a lot of people stopped eating beef. Should we remove the beef from the CPI basket?

Similarly, the mortgage rates skyrocketed, and the number of home purchases plummeted. But many more people would buy a home, if only they could afford it. Just like more people would start eating beef again, if they could afford it.

>How do interest rates effect everyday people exactly, other than price inflation on goods and services (which is included separately in CPI)?

We artificially made interest rates low (look up open market operations). Low rates make risky ventures more financially attractive by making the DCF denominator smaller. People look to invest in growth instead of reliable revenue. That is, "bet on the future" becomes dramatically more attractive than "goods and services being made now". Additionally, it devalues wages and increases the value of financial assets. It's literally "rich get richer: the policy".

Anyone who is upset about NFTs millionares existing while EMS workers and teachers struggle, or upset about billionaires' staggering wealth inequality, or fraudulent do-nothing scam businesses like WeWork and Theranos, obviously stupid ventures like Juicero, or basically any economic upside-downness that most laypeople have recently come to blame on "capitalism" need look no further than LIRP and ZIRP. We snapped all the fingers of the invisible hand, but did it far upstream of anything that average people pay attention to. Anyone who wasn't paid in equity got fucking robbed over the last 20 years.

I don't really agree that interest should be included in CPI, but I do agree that CPI (and PCE) is an absolute joke that doesn't measure what it claims to.

What is confusing? The argument for including them is that people pay them.

>Blindly stacking

Stawman nonsense that literally no one suggested

Reply on news.ycombinator.com