The rent & vacancy numbers you see from Eudaly et al aren't Real

Just a few things that distort the rent levels you see in the news etc.

Most apartments aren't rented in a given cycle.  Rent levels I have seen quoted are based on current listings rather than a survey of existing renters.

In surveys I have participated in myself, the question is always what I am currently renting, say, a 1 BR for and not  what my average rent is on all my 1 BR units.  

Actually, a good metric for this number would be to ask what average rents are for a given manager's one bedrooms vs what current listings are going for.

Of course, current listings are further distorted by new construction.  New construction (even the huge amount we've had lately) is a relatively small percentage of the overall rental stock.

But it is a much much larger percentage of the apartments currently listed and and even larger percentage of the apartments currently vacant (since in lease ups, you only list a couple of sample units).

So that makes current listings a pretty poor proxy for rent levels in the City over all.

I haven't seen anyone with statistical knowledge tackle this stuff.  PSU reports have devolved into meaningless quotes of the surveys that are better read in the original (like the housing association surveys & Barry Report). 

Marcus & Millichap has the numbers, but they skew them for marketing - depending on whether  they're talking to a seller or buyer.   

Sorry for the rant! But you'd think with something as serious to the City as vacancies and rent levels - especially in the last year or so, someone would really try to figure out meaningful numbers.  The hassle is that the current numbers serve the political ends better than real numbers or numbers actually calculated for the purpose of policy. 

Personally, I'd like to see something like Same Store Rents.  E.g What is the same apartment renting for now vs its last rental.  By analyzing a significant number of apartments that way and given the dates between the leases, you can extrapolate year on year changes.  You also filter outliers like new construction and buildings that have been significantly improved.   That is exactly how the CoreLogic Case-Shiller index works.   The current quoted stats are more like what you see from the National Association of Realtor index - based on recent sales.   

Anyway, a lot of this is just based on my experience of seeing rents sometimes dropping a lot on my units while the media reports rents leaping up and up.  Though my rents, as you'd expect for this time of year, have been rising.

(Seasonal adjustment! Another thing we should see in quoted numbers that aren't comparing year on year).