# Is paying mortgage points a good idea?

A couple I know are considering whether to pay mortgage points and this got me thinking. Their loan amount is \$540,000. If they put in \$10,800 (2% of the loan amount) for the points, then the mortgage rate reduces from 3.5 to 3.125. This saves around \$40,000 for a 30 year mortgage. I have read that paying points is not a good idea if one is not planning to live in that house for a long time (I mean even before the amount you pay in points breaks even).

Question:
Let's say if they are going to live in that house for 30 years, in that case would paying points still be a good idea. Like my thinking is if I were to invest the \$10,800 where the interest rate is a modest 6% (this is modest I suppose, usually its 8 to 10, right?), then the returns are \$62000 after 30 years. Given this scenario, paying points doesn't seem to be a good thing. Would be curious to know if my thought process is right here.

(I know I need to consider inflation after 30 years too, but I am unable to see how should I compare the \$40,000 (from interest savings) to \$62000(investment return) with respect to inflation and which is better in that case. Any ideas?)

• Don't forget that if this is a new mortgage, the points are tax deductible. If it's a refinance, you divide the amount by the number of years of the loan and deduct that amount each year. Aug 28, 2016 at 18:20
• @mkennedy - excellent reminder, as that changes the math by 25% or more for this situation. Aug 28, 2016 at 18:38
• "where the interest rate is a modest 6% (this is modest I suppose, usually its 8 to 10, right?)" - That sounds overly optimistic to me. What would you invest in for that return? Aug 28, 2016 at 20:16
• Great point @mkennedy Aug 29, 2016 at 2:39
• I did the calculations again after this new info about tax deductions and found that there is not much difference between buying points now and investment return on a 6% interest rate. As @Aganju's answer mentions probably buying points is at an advantage for the bank otherwise why would they do it. For example they could charge more than 6% when giving loans which is a good for them. Thoughts? Aug 29, 2016 at 2:46

Two things are to be considered:

1. You need to look at the points in time when those cash flows become real. Having \$40,000 after 30 years is not the same as having \$40,000 now. To compare the options realistically, you need to convert all payments to the same point in time, by multiplying (or dividing) them with the assumed inflation ratio for each year: let's say you assume 3% inflation, then next year’s \$1,000 payment would equal \$1,000/1.03 = \$970.87 this year; etc.

2. If you refinance after 5 years or after 10 years, the money you paid on the points is (partly) wasted, as you will not have the advantage of the lowered rate for the future. Can you ever be sure that interest rates will not go lower for the next 30 years?

Considering those two points, it is very difficult to say if points are a good idea or not. Typically, it evens about out for average behavior - that is how the bank calculates their offer, after all - but often enough it is to your disadvantage, as it locks you to that bank and mortgage.

It also could be a good strategy to take 'negative' points - meaning you get extra cash-out but sign with a higher interest; you can then use that extra money to pay off some principal right away a week later, and refinance out of the higher interest rate as soon as possible (of course you can also blow the extra money for that new 150" color TV, which is what the bank would love you to do, but that is not a good strategy). Do your own math!

• Mostly agree. But, to your point 'b', one can calculate a breakeven time, so it's not a risk of lower rates during the 30 years, but between now and that breakeven date, about 6 years hence. Aug 28, 2016 at 18:15
• Ordinarily I would discount at the interest rate of the loan, not the inflation rate. Wouldn't you? Aug 28, 2016 at 21:10
• Why assume almost 3% inflation when we haven't had that for years, in a deflationary basic environment that's likely to continue until the bulk of the Baby Boomers have died off? Aug 29, 2016 at 14:23
• @MasonWheeler Interestingly was just reading an article in Bloomberg (I think) about how people under 40 have no experience with inflation and therefore expect that there will be none in the future and how this is probably wrong. I'll try to find it. Aug 29, 2016 at 15:55
• @MasonWheeler Sorry, still not making sense. When they spend less it's bad but when they spend nothing that's good? Aug 29, 2016 at 18:47

It's easy to overanalyze these scenarios. Simpler is better.

They are looking at paying 2% to gain 3/8%/year. This is 5-1/3 years to break even. Since a bit of return is lost over those 5 years, adjust, and call it 6. That's it. 6 years to break even. If they sell or refinance in less time, they lost a bit of money. If they stay put, they start to gain after 6 years.

Note - you can write a spreadsheet and compare the payments and balance remaining over time to see the exact cost of early refi/sale vs gain over time. Just be sure to account for the expense of the points, e.g. as a side investment, or prepayment on the loan. And for the next level of complexity, account for the tax adjustment (the deductibility).

• so, 6years to break even, over a decade to double your money? seems like investing in stocks would be a better deal. Aug 29, 2016 at 17:34
• Not really. You are welcome to author another spreadsheet. Different from mine. In your case, take the \$10,800 and 'invest' on the side with whatever return you wish. For the point payer, take the monthly savings and do the same. You'll find that as you 'turn the dial' on returns, a very high rate of return will make the break even on points longer than 30 years, or never. In my spreadsheet below, the effective rate for the exercise is the mortgage rate itself. A double at 3.5% is about 20 years. Aug 29, 2016 at 18:45
• I admire your desire to add this additional layer of complexity, but it's exactly what my answer above cautions against. It spills into the "pay mortgage early vs invest" debate, which this question doesn't really need. Aug 29, 2016 at 18:46

Going against my own answer regarding overanalyzing, this is the spreadsheet version. Of course, some explanation is required. Two adjustments made to keep the comparison apples to apples. First, I needed to account for the no-point scenario to have kept their \$10,800. I applied it to the balance, so you can see the mortgage balance drop by that amount beyond the regular payments. Next, the payments for the lower rate loan would of course be lower, but we need to account for that monthly difference, so for the point loan I use the same payment as the no-point loan. This shows a running difference between the payoffs, and you can see the breakeven occurs in year 6.

I completely ignored taxes in an effort to keep the math simple. If I make one change, assume a 25% rate, and see that the points don't cost \$10,800, but a net \$8,100, and apply just that to the no-point loan principal, we see below, the breakeven before the end of year 5.

It's tempting to get carried away. Use the saved points (or saved payment for the point payer) to invest at some higher rate. Or to tax adjust for all payments until the breakeven point.

Even with all the numbers, the question is whether the buyer will stay in the house, and stay with the mortgage more than X years, and these spreadsheets show 5-6 years is about the right timeframe.

To make a valid comparison between two scenarios, you need to do two things:

1. Make the two cash flows over time for the two scenarios as equal as possible;
2. Never simply add amounts of money paid or received at various times.

Of course, having a flexible mortgage calculator is a basic necessity as well. One that allows you to set or calculate any of principal, regular payment, number of payments, balance owing, and interest rate would be nice.

First, lets consider paying the full mortgage amortization:

Mr. Points borrows \$550,800.00 at the reduced rate of 3.25% compounded monthly for 30 years of monthly payments. He is committed to paying \$2397.12 monthly for 30 years, and after paying \$10,800.00 for the points, has \$540,000.00 to do with as he wishes on the day the deal closes.

Ms. NoPoints (by #1 above) also agrees to pay \$2397.12 monthly for 30 years at the original rate of 3.5%. It turns out that this payment schedule and interest rate will support a loan of \$533,826.60, all of which Ms. NoPoints can use as she wishes.

So, on the day of the loan, Mr. Points is ahead by \$6173.40. Clearly, take the points!

Now, consider that the mortgages will be paid off after 10 years.

Exactly as before, Mr. Points borrows \$550,800.00 at the reduced rate of 3.25% compounded monthly for 30 years of monthly payments. He is committed to paying \$2397.12 monthly for 30 years, and after paying \$10,800.00 for the points, has \$540,000.00 to do with as he wishes on the day the deal closes. However, after making these payments for a while, he pays off the mortgage in full after 10 years, when the balance owing is \$422,625.43

Ms. NoPoints insists on exactly this same cash flow; paying \$2397.12 monthly for 10 years while the debt grows at 3.50% compounded monthly, and then settling with a lump-sum payment of \$422,625.43 It turns out that this cash flow and interest rate will support a loan of \$540,386.66

So, under this scenario, Ms. NoPoints is ahead on the day of the loan by \$386.66

• What the factorial? On day one, the point payer paid \$10,800. And the non-payer is ahead \$10,800. The future hasn't occurred yet, despite the 3 TV shows containing a time travel premise coming to US TV this fall. You can do math looking at the monthly savings the point payer has, and view it from a number of angles, mine was just one way. You might even prove mine lacking in a few regards. But the day of the loan, there's no question, one is ahead exactly \$10,800 and over time that changes. Aug 29, 2016 at 13:48