[rory]

Average market rent represents a segment of rent prices that responds much more quickly to market price shocks than rent prices as a whole. Your link is effectively the price change for people moving into a new apartment. The other 50-60% of the market (people renewing their lease or continuing tenancy at will) has much smoother price growth over time.

Doesn't square 17% and 3%, but it could definitely mean the lagging CPI rent growth we see over the coming year is more like 9 or 10%.

[adam_arthur]

The market rate forecasts the rent increases for the broader market.

If market rate of new units is 20% higher, almost by definition you'll see 20% higher rents for all in the long run.

Why should we use backward looking methodology in computing the CPI? It's an extremely flawed way to analyze the state of the market. Forward looking metrics are more useful to policy makers.

The methodology for OER and rent analysis in the CPI acts to suppress true price increases. Though those price increases eventually materialize as people renew their leases, move etc. It comes with a year or longer lag though.

The Fed drives their policy based on this data. Why should the data have a one year lag and potentially put them behind the curve?

If we used the same CPI formula as in the 70s, we would be seeing similar CPI numbers too.

[rory]

> If market rate of new units is 20% higher, almost by definition you'll see 20% higher rents for all in the long run

Well, yes, but that's not very meaningful without knowing over what time period you'll see that increase.

> Why should we use backward looking methodology in computing the CPI?

I mean that's literally just what CPI is. You're welcome to use leading indicators if you want, and economists certainly do. One slight problem with predicting the future is we don't know what will happen then.

> If we used the same CPI formula as in the 70s, we would be seeing similar CPI numbers too.

This is completely unsupported by the data. Yes, the switch to geometric mean effectively deflated CPI when compared to measures taken using the previous formula, but not remotely close to a level that would bump our current ~6% rate into the high teens.

[adam_arthur]

Home prices used to be included in the CPI formula, which is how you get to similar number.

Not OER, but actual home prices.

But even if CPI reflected rents fully, today, we would be at similar numbers.

And the Fed primarily looks at core PCE which is based on the backward looking CPI data. It's very obvious from a policy perspective, they should be considering market rate and not existing tenant rates for forecasting forward inflation.

Yes, the Fed forecasts the backward looking CPI, but their forecasts were also so far off this year, I wouldn't put any stock into them. I believe they predicted 3% inflation for 2021 earlier this year.

[rory]

> Home prices used to be included in the CPI formula, which is how you get to similar number.

Sorry, I thought you were making the geometric mean argument. Yeah that would definitely add significantly for this year in particular. Home price inflation is about 20%, and is replacing housing inflation at maybe 3% in the CPI calc. Owner occupiers are ~64% of the housing market which contributes around ~33% to CPI. So including that would add (.2 - .03) x .33 x .64 = .036 so 3.6% to CPI.

So around 10%. Comparable to most of the 70s, but not the high-teens emergency points.

And yes the Fed doesn't always accurately predict inflation, but in the long run no one else does either (and one could make billions doing so, it's not like no one's trying).

[Lhiw]

> So around 10%. Comparable to most of the 70s, but not the high-teens emergency points.

Except this is not comparable to most of the 70s at all.

Household leverage and disposable income change how serviceable debt is at different rates.

The numbers I saw puts a 2021 mortgage rate of 3% at an equivalent benchmark rate "in your pocketbook feels" as a 6% rate in pre gfc 2007.

Are you sure 10% doesn't warrant the same emergency as the high-teens of the 70s?