Source: Quadlet

A hat tip to John Keith over at [The Boston Real Estate Blog](http://bostonreb.com/blog/?p=1067) who posted this link [Are Real Estate Prices Dependent on Mortgage Rates? [Quadlet Consulting]]) and asked for my feedback. Actually I have also seen this study linked to a bunch of other sites but felt that it was so weak I was reluctant to post about it because I didn’t know where to begin.

The study stated that by looking at Manhattan average sales price and correlating it to 30-year fixed mortgage rates, it is inconclusive that mortgage rates influence values.

The study uses data from the Manhattan real estate broker Corcoran Group that goes back to 1974. I find this resource interesting because Corcoran was founded in the late 1970’s or early 1980’s and therefore it is not clear where their data came from prior to that unless it is a very thin set from early on and later data is based mainly on their own sales. However, the trend line does appear reasonable.

I think a lot of the problem is the data set chosen. He uses annual data provided by a brokerage firm. The subject is the Manhattan market which does not always behave like the national market due to the international nature of the local economy as well as the predominant co-op housing stock (this has kept investors out during this recent boom). He might have considered NAR or OFHEO stats since his mortgage rates are national. Also, he does not use quarterly data (Corcoran doesn’t have it), so its not going to appear very responsive in graphic form. He uses fixed mortgage rates yet adjustable rates are one of the primary drivers of the recent housing boom. Plus he makes the assumption that mortgage rates are the only factor that causes prices to fluctuate. Over the past 10 years, the Manhattan housing market has been plagued by the chronic limitation in supply. So the premise of rising rates causing prices to slow would not be immediate if inventory is tight.

Despite all this, my stats do show a clear pattern of influence of mortgage rates using a [CPI adjusted 1-year adjustable](https://millersamuel.com/charts/gallery-view.php?ViewNode=1136992267ggIva&Record=6) and a [30-year fixed unadjusted](https://millersamuel.com/charts/gallery-view.php?ViewNode=1067010393ySSAL&Record=5).

I think the bottom line here is that the issue with drawing this sort of conclusion, its that it seems to be based on the quality of the graphics, and not much more than that.

I would have to conclude that his effort appears noble but it is overly simplistic and misleading.