I have to disagree with you that you can see where rates are going in the future from the trend in the recent past. Or perhaps I should say that I can’t! If you can, then I suggest you should trade bonds and you could make many, many times your current earnings. (I’m not being sarcastic.)
I suspect you don’t actually advise your clients that rates are more likely to go up before lock-in, or vice versa. I think the consensus view amongst interest rate experts on what anyone can safely and accurately say about future long rates, and hence the lock-in (fixed) rate is that:
1. The best estimate of it is roughly the current rate
2. Rates will go up (or down) by an uncertain amount from now until lock-in
3. The degree of uncertainty increases with time to lock-in
(and for nerds, it’s approximately proportionate to the square root of the time to lock-in, if the time is short enough)
4. The degree of uncertainty varies with the amount of daily volatility in the rate today (and for nerds, it’s directly prportional over short enough times to lock-in)
Here’s a practical example:
Your buyer is trying to decide between two purchases. Home A can be locked in 1 month from now; B 4 months from now. Otherwise, they are equal. Current fixed rates are 6%. The buyer wants to compare the interest rate risk. He decides to look at the cost of home A at 6.5% as a reasonable worst-case, 50bp higher than today’s rate. Then it’s fair to look at home B at 7% [twice as much over today’s rate, because sqrt(4 months/1 month) = 2].
Here’s another:
Same buyer, but now the daily fluctuations in the rate have gone up by 50%, and the buyer is worried his 6.5% one month from now isn’t worst-case any more. How high should he go? 6.75%. Pretty obvious.
Sorry about all that math, and the absence of the excitement of making a bet on rates rising or falling. Maybe it’ll be of some practical use to you or someone else.
