5 ms·

CPI is fine...

by ivalm 6y ago

CPI is fine...

alasdair_ 6y ago
CPI is also a measure that is altered over time by groups with a vested interest in keeping “inflation” low. As a very simple example of why this may be misleading, consider that house prices are not considered in the “basket of goods” that Americans buy, despite this being literally the most expensive purchase most people ever make. The formula itself is also changed from time to time. If we were using the “old” formula, inflation would be MUCH higher (shadowstats.com for current numbers).
- perl4ever 6y ago
  >house prices are not considered in the “basket of goods” that Americans buy This is (widespread) mis/disinformation. The CPI carefully distinguishes between houses as capital goods and shelter/housing as something people consume. If you Google it, you will find that the equivalent rent for homeowners is incorporated in the CPI. Even if they ignored homeowners, the index would be very little changed if it was based on renters alone. See https://www.bls.gov/cpi/factsheets/owners-equivalent-rent-and-rent.pdf https://www.bls.gov/cpi/factsheets/owners-equivalent-rent-an...
  - alasdair_ 6y ago
    I’m aware of that. The problem with equivalent rent is that it’s seldom actually equivalent. In order to compute the CPI-U, the people doing the work essentially call up 14,000 people and, if they own a home, they ask them how much they would theoretically charge in rent if they rented their home out. The problem with this is obvious - most people do not do rent income estimates for a living and so the numbers are often quite far from reality. More importantly, people find it very difficult to estimate future interest rates and they fail to take into account things like the sunk cost fallacy (see: 2008 financial crisis, where people in crashing markets still guessed they would be able to rent their homes based on pre-crash prices). There is also the issue of a negative feedback loop, where low interest rates lead to lower than expected inflation, because “house prices only every go up baby!” - or in other words, people are perfectly willing to accept $3000 in rent even if the mortgage is $5000 if the house is increasing by $8000 a month in appreciation. This appreciation is hidden from the CPI because only theoretical rent is considered. The net result of the negative feedback loop is the fed sees that inflation is low, so it thinks it can keep rates low too, which leads to asset bubbles like in 2008. (And, likely, in 2020 with stocks too...)
    - perl4ever 6y ago
      You're saying sometimes people estimate equivalent rent too high and sometimes too low. How is that an argument that inflation is systematically underestimated, let alone that home prices are ignored? Separating housing as an investment from housing as consumption is obviously the right thing to do. But if it wasn't separated, it doesn't make sense it would add systematic bias in one direction. Prices go up, prices go down. Inflation would be pushed higher and then lower, relative to the "true" value.
      - cameldrv 6y ago
        Perhaps, but it's also not a good measure of the market. When the BLS wants to measure the price of other consumer goods, it goes to a store and sees how much something costs. When goods change, it makes quality adjustments. This would be the equivalent of walking down the street and asking people "how much can I pay you for your jacket?" It's a very strange way of measuring the price for something when there is a perfectly well functioning rental market.
        perl4ever 6y ago
        > It's a very strange way of measuring the price for something when there is a perfectly well functioning rental market. They do survey people who rent as well as own, so they are measuring the market directly for that portion. As far as the owners go, I think you're disregarding the fact that the specific houses that are not being rented do not have more authoritative rents. If you use actual rental data, then you have to have a model using assumptions to map one population to another, which seems fraught with potential problems. The point of a survey is to collect survey data, and if they don't do it, then it is that much worse of a survey, because there is no 100% substitute.
        alasdair_ 6y ago
        >The point of a survey is to collect survey data, and if they don't do it, then it is that much worse of a survey, because there is no 100% substitute. Not 100%, but I bet doing something as simple as looking at the bank accounts of the selected groups in aggregate (over a long time period) and adding in things like tax returns could get quite a bit of the way there. To be honest, I'm surprised something like mint.com or Personal Capital don't offer their own interest rate measures.
- imtringued 6y ago
  The problem is that consumer prices aren't rising. Right now whatever the central bank/government is doing is only affecting mortgages, student loans (more of a government thing) and stocks (through QE). None of the money is actually going into the "real" economy.

CPI is fine...

by ivalm 6y ago

CPI is fine...

CPI is also a measure that is altered over time by groups with a vested interest in keeping “inflation” low. As a very simple example of why this may be misleading, consider that house prices are not considered in the “basket of goods” that Americans buy, despite this being literally the most expensive purchase most people ever make. The formula itself is also changed from time to time. If we were using the “old” formula, inflation would be MUCH higher (shadowstats.com for current numbers).

perl4ever 6y ago
>house prices are not considered in the “basket of goods” that Americans buy This is (widespread) mis/disinformation. The CPI carefully distinguishes between houses as capital goods and shelter/housing as something people consume. If you Google it, you will find that the equivalent rent for homeowners is incorporated in the CPI. Even if they ignored homeowners, the index would be very little changed if it was based on renters alone. See https://www.bls.gov/cpi/factsheets/owners-equivalent-rent-and-rent.pdf https://www.bls.gov/cpi/factsheets/owners-equivalent-rent-an...
- alasdair_ 6y ago
  I’m aware of that. The problem with equivalent rent is that it’s seldom actually equivalent. In order to compute the CPI-U, the people doing the work essentially call up 14,000 people and, if they own a home, they ask them how much they would theoretically charge in rent if they rented their home out. The problem with this is obvious - most people do not do rent income estimates for a living and so the numbers are often quite far from reality. More importantly, people find it very difficult to estimate future interest rates and they fail to take into account things like the sunk cost fallacy (see: 2008 financial crisis, where people in crashing markets still guessed they would be able to rent their homes based on pre-crash prices). There is also the issue of a negative feedback loop, where low interest rates lead to lower than expected inflation, because “house prices only every go up baby!” - or in other words, people are perfectly willing to accept $3000 in rent even if the mortgage is $5000 if the house is increasing by $8000 a month in appreciation. This appreciation is hidden from the CPI because only theoretical rent is considered. The net result of the negative feedback loop is the fed sees that inflation is low, so it thinks it can keep rates low too, which leads to asset bubbles like in 2008. (And, likely, in 2020 with stocks too...)
  - perl4ever 6y ago
    You're saying sometimes people estimate equivalent rent too high and sometimes too low. How is that an argument that inflation is systematically underestimated, let alone that home prices are ignored? Separating housing as an investment from housing as consumption is obviously the right thing to do. But if it wasn't separated, it doesn't make sense it would add systematic bias in one direction. Prices go up, prices go down. Inflation would be pushed higher and then lower, relative to the "true" value.
    - cameldrv 6y ago
      Perhaps, but it's also not a good measure of the market. When the BLS wants to measure the price of other consumer goods, it goes to a store and sees how much something costs. When goods change, it makes quality adjustments. This would be the equivalent of walking down the street and asking people "how much can I pay you for your jacket?" It's a very strange way of measuring the price for something when there is a perfectly well functioning rental market.
      - perl4ever 6y ago
        > It's a very strange way of measuring the price for something when there is a perfectly well functioning rental market. They do survey people who rent as well as own, so they are measuring the market directly for that portion. As far as the owners go, I think you're disregarding the fact that the specific houses that are not being rented do not have more authoritative rents. If you use actual rental data, then you have to have a model using assumptions to map one population to another, which seems fraught with potential problems. The point of a survey is to collect survey data, and if they don't do it, then it is that much worse of a survey, because there is no 100% substitute.
        alasdair_ 6y ago
        >The point of a survey is to collect survey data, and if they don't do it, then it is that much worse of a survey, because there is no 100% substitute. Not 100%, but I bet doing something as simple as looking at the bank accounts of the selected groups in aggregate (over a long time period) and adding in things like tax returns could get quite a bit of the way there. To be honest, I'm surprised something like mint.com or Personal Capital don't offer their own interest rate measures.
imtringued 6y ago
The problem is that consumer prices aren't rising. Right now whatever the central bank/government is doing is only affecting mortgages, student loans (more of a government thing) and stocks (through QE). None of the money is actually going into the "real" economy.