Friday, August 26, 2016

Some problems I have with tying owner equivalent rent to CPI

Unfortunately I don't know the nuts and bolts of how interest rates got incorporated into owner equivalent rent (OER) in the 70s, so I'm not sure I understand CPI and OER.

But someone (Dean Baker?) has written about an interesting thing:

Apparently, sometime in the 70s the Fed changed the way it calculated CPI: it started to include change in mortgage interest in OER.

Obviously, that would have a tremendous impact on CPI readings later that decade: the Arab oil embargo, and later the Iranian revolution, generated massive short-term commodity-driven spikes in inflation.

But Dean Baker (or someone whose paper he came across, I dunno) has apparently gone into the inflation data and found that, if you back out the inflation/interest component of OER, there actually was no inflation problem in the 70s aside from those two supply shocks.

As I said, I don't know enough about the mechanics of OER and CPI calculation, and I can't find the paper (if there is one) that explains this stuff in detail. But own my own I can see massive methodological problems with incorporating inflation into CPI:

1) if you take the aggregate mortgage costs as changing with interest rates (debatable, see 2 below), then the NPV of future payments with which you might calculate OER is only known post-hoc: you can't extrapolate NPV from the present rate, because the present rate will continue to change, and compounding can introduce almost an order of magnitude of error! So it depends how OER amortizes this incredibly indefinite NPV.

2) In my own parent's case, they were locked in at 6.25% all through the 70s, so couldn't you say their OER went *down* in real terms? Basically the Fed would have to incorporate into its calculations some measure of how many mortgages are locked-in, and then (for the locked-in crowd) adjust their OER for future increased wealth (see 4 below).

3) Mortgage interest, in the US, is tax-deductible anyway, so wouldn't that make the delta of OER deductible?

4) Housing's an asset, so if rapid house price accumulation eats into current consumption, it'll also radically increase future income (in something like a sophomore economics 2-period everybody-dies model). Essentially, OER is paying for an asset (and theoretically thus paying for future income): it's not consumption. Essentially the home owner is getting that money back someday.

5) And does an interest-adjusted OER also take into account how increasing home prices increase the wealth of home owners with no mortgage?

6) If we incorporate price inflation of housing assets into CPI, why not incorporate price inflation of other assets, like stocks?

7) Similarly, if interest rate increases inflate OER, does the Fed also take into account the effect of interest rate increases on the NPV of firms' capital? I.e., is corporate rented-to-own capital treated the same in CPI?

No comments:

Post a Comment