5

I was reading about the calculation of the Shelter index of the CPI which accounts to more than 30% of the entire basket. It consists mainly of two parts: (1) Rent of primary residence, and (2) Owners’ equivalent rent of residences (OER).

On how OER is obtained the BLS writes:

The expenditure weight in the CPI market basket for OER is based on the following question that the Consumer Expenditure Survey asks of consumers who own their primary residence:

“If someone were to rent your home today, how much do you think it would rent for monthly, unfurnished and without utilities?”

It just feels odd on the face of it: why bother with the subjective/sentiment "equivalent" value of the EOR instead of looking only at the actual rental data from real contacts?

In theory, in absence of subjectivity, shouldn't they be equal?

d_e
  • 283
  • 1
  • 6

1 Answers1

4

There are host of reasons:

  1. Rental houses are not randomly allocated across country. It is very likely that most rental houses are not directly comparable to non-rented houses. Of course sometime you can find 2 exactly same houses in the same neighborhood where one is rented and other not, but that is very rare.
  2. People who rent houses are different from people who don’t rent houses. Value is purely subjective as understood by modern economics. If someone is not renting home at a current market price then that means the current market price is simply too low for that person to rent. That is objective revealed preference of the person not renting. So we can a priori say it is likely that person requires higher rent to rent (even though we don’t know what the higher rent would be).
  3. While asking the person is not perfect (to get the objective price it would be better if the statistician would pretend they actually want to rent a room/house and then record negotiated price) it is much better than just taking price of some non-comparable rented house from non-comparable individual given the points 1 and 2.
1muflon1
  • 56,292
  • 4
  • 53
  • 108