next »

[pedanterast]

http://money.msn.com/saving-money-tips/post.aspx?post=7eb8f8e1-7326-4d33-b1ef-74b1fe01dff2

The gist of the article is that the "average" 30-year fixed mortgage rate has dropped below 4%, and refers to a spate of re-financing.  I see two issues with the article (grammar and failure to spell check aside).  Number 1 is that the article refers to "points."  Well, if they are paying 0.7 points to "get 4%," they aren't getting 4%, are they?

Number 2, it refers to an "inpressive" [sic] chart in the Christian Science Monitor.  This chart is particularly unimpressive to me; in fact, it's misleading and inherently defective, because it does not go through the origin and thus artificially overstates the amount of change.

[aandsdean]

http://money.msn.com/saving-money-tips/post.aspx?post=7eb8f8e1-7326-4d33-b1ef-74b1fe01dff2

The gist of the article is that the "average" 30-year fixed mortgage rate has dropped below 4%, and refers to a spate of re-financing.  I see two issues with the article (grammar and failure to spell check aside).  Number 1 is that the article refers to "points."  Well, if they are paying 0.7 points to "get 4%," they aren't getting 4%, are they?

Number 2, it refers to an "inpressive" [sic] chart in the Christian Science Monitor.  This chart is particularly unimpressive to me; in fact, it's misleading and inherently defective, because it does not go through the origin and thus artificially overstates the amount of change.

Well, yeah, they are getting 4% by prepaying a very small amount of interest at the front end (.7%, one time) to lower the interest over the life of the loan, which increases the APR by--get this--.0023%.

So, I guess that technically the APR goes to 4.0023%.  It's a rounding error.

You are so intent on proving others stupid/foolish/innumerate that you have a drastic tendency to overreach rhetorically. 

[pedanterast]

But your math is only good if they keep the mortgage for thirty years.  Anyway, I got a 1.5% difference, calculated as follows, using a hypothetical $100k principal:

FV = 0; PV = 100,000; N = 360; I = (4/12), then compute the payment:  $477.42

Then use that payment, subtract the 0.7% in points, making the PV (net loan proceeds) 99,300, and recompute I: (.338...x 12) = 4.05827 ....

Then divide that by 4 and subtract one:  .0145676 .... , or slightly less than one and one-half percent difference in the rates. If you use monthly compounding you would get .0147709 ... , which is more accurate.

I understand you are talking about percentage points versus the percentage difference, but I still don't see how you got 4.0023%.  Care to share your methodology?  Was that by amortizing the points over the life of the loan instead of taking them out up front?

The chart bothers me more, of course.  But the mortgage companies should have to state the effective annual rate including all the closing costs, in my opinion.

Original "30 year" mortgages have an actual life of about seven years (which is one reason the points get taken out up front), so the true effect of the up-front points will be considerably greater for most borrowers. 

 

[galactic_hedgehog]

Number 2, it refers to an "inpressive" [sic] chart in the Christian Science Monitor.  This chart is particularly unimpressive to me; in fact, it's misleading and inherently defective, because it does not go through the origin and thus artificially overstates the amount of change.

The origin?  I don't think they had mortgages back in the year 0.  Or do you mean you think it would be better to show the changes in interest rate as a percentage relative to some specific date/rate?  Potato, potahto.

[pedanterast]

I mean the Y-axis should start at zero.  The X-axis can start wherever the data started. Of course, that could be construed as data mining (why pick 2007?) but if the chart is labeled with the correct years I don't see an issue there).  I get what you are saying; I misused the term "origin."  The X axis isn't expressed terms of numbers but rather in years.  I'd probably have two labels on the X-axis, years and "n" number of periods.  That would be helpful, I think.

[totoro]

It's pretty impressive to me. It shows that mortgage rates almost halved over this period.

[pedanterast]

You're a sophisticated user.  To Joe Six Pack it visually could well appear that mortgage rates are approaching zero.  It's like lining up tinnies against stubbies.

[afm_man]

They could have started the chart at 1980 and then it would really look impressive.