bigwilly Wrote: ------------------------------------------------------- > think about it. You have different levels of > convexity. CSK you’re the Math genius of the > bunch you should know this 
Convexity >0 is a > positive. lol, no i know that convexity can be 5 or 4, but i was just wondering why it would be more convex, i guess i see that point that since option still have some value (even deep out of the money) you will get beneift from option price decrease
PJ - a noncallable would b/c the prepayment option decreases the overall bond price more than just a decrease in itnerest rates alone would suggest.
Okay… So here’s my summary that I"m sticking with and hoping it’s enough: - When current interest rates (yield) is above the coupon rate of a MBS, the pre-payment option on the MBS is out of the money and the MBS acts like a regular non-callable bond with positive convexity. - When current interest rates (yield) is below the coupon rate of a MBS, the pre-payment option of the MBS is in the money and the MBS will exhibit negative convexity and hence will not benefit as much from a further drop in rates because the pre-payment option value will rise and mitigate some of the gains in the price of the MBS. As a result, Non-Callable bonds will outperform MBS when current yields < coupon rate of the MBS. When current yields > coupon rate, the performance of a MBS will be in line with that of a non-callable bond. As a side note, the implication of positive convexity is that as interest rates continue to fall, prices increase at an increasing rate because duration is getting longer (slope of the tangent line). The opposite is true in negative convexity. Do I have the above information correct?
as interests rate fall your duration would decline (or am I totally out of line???). recall LVL 1 - when you were asked to argue which bond wud have lower duration - 7% 10 yr bond or 5% 10 year bond (7% would be the answer). also positive convexity cushions both upside and downside effects of duration. notice it is in opposite direction to duration in calculation of price (- duration*bps change + 0.5*convexity*(bps change^2))
I thought the reason callables outperform non-callables when int rates increase is because they have a higher yield to start with to make up for the fact that they have to endure having the bond called at lower rates, therefore when rates rise they have the same positive convexity as a non-callable, but they simply outperform because they have that extra yield baked in?
Yes that is true, BUT the option also helps give it a bit of a boost.
PJStyles Wrote: ------------------------------------------------------- > Okay… So here’s my summary that I"m sticking > with and hoping it’s enough: > > - When current interest rates (yield) is above the > coupon rate of a MBS, the pre-payment option on > the MBS is out of the money and the MBS acts like > a regular non-callable bond with positive > convexity. > > - When current interest rates (yield) is below the > coupon rate of a MBS, the pre-payment option of > the MBS is in the money and the MBS will exhibit > negative convexity and hence will not benefit as > much from a further drop in rates because the > pre-payment option value will rise and mitigate > some of the gains in the price of the MBS. > > As a result, Non-Callable bonds will outperform > MBS when current yields < coupon rate of the MBS. > When current yields > coupon rate, the performance > of a MBS will be in line with that of a > non-callable bond. > > As a side note, the implication of positive > convexity is that as interest rates continue to > fall, prices increase at an increasing rate > because duration is getting longer (slope of the > tangent line). The opposite is true in negative > convexity. > > Do I have the above information correct? 100% correct. that is the way i’ve always looked at it (though i’m not sure it’s as simple as saying that the coupon rate is the determining factor). my guess if we get an example it will state interest rates are either very high or very low. anything else and it’s a bs question.
I disagree slightly, let me know your thoughts. MBS/Callable = long bond + short call option when rates are low, call option has lots of value and thus MBS price drops when rates rise, call option value approaches 0, it does not turn negative-option values are never negative, it simply does not have value, leaving you with the value of the long bond, which for a MBS is higher as it compensates for possibility of security being called?
sdcfa Wrote: ------------------------------------------------------- > I thought the reason callables outperform > non-callables when int rates increase is because > they have a higher yield to start with to make up > for the fact that they have to endure having the > bond called at lower rates, therefore when rates > rise they have the same positive convexity as a > non-callable, but they simply outperform because > they have that extra yield baked in? yup. the way i’ve looked at it is that because of the call option they are initially priced lower (higher yield) and when rates rise the call option becomes less valuable (or the bond is less likely to get called) and therefore callable bond prices “catch up” to non callables (because they are now basically priced as noncallables) and therefore will outperform noncallables from hte issue date. they would not outperform if rates rose when they were already high enough that the bond wasn’t likely to be called. my .02
Yes But, as teh Option Value Decreases to 0, I believe it gives it a boost… For example If Year 0 Treasury Bond = 100 Call Option = 5 MBS = 100 – 5 = 95 Now Interest rates Increase say 5%. Now lets just say these are the new hypothetical prices Treasury Bond = 95 Call Option = 2 MBS = 95 – 2 = 93 Now what was the Performance of each Treasury = 95/100 = -5% MBS = 93/95 = -2.11% So MBS performed better b/c of the Decrease in the Value of the Call Option.
agreed bigwilly, just that it’s posibible for that outperformance to be very small. With MBS wouldn’t they underperform treasuries in a rising rate enviornment because part of the reason investors want MBS is because they get a % of principle back because of prepayments. If rates rise, prepayments drop, therefore MBS become less valuable because you are getitng your principle back later than expected.
can we please get confirmation as to whether MBS outperformance non-callable bonds when the pre-payment option is in the money and there is a rate increase ie: interest rates are currently low relative to the coupon rate and then there’s a subsequent increase in rates. My understanding is that yes you gain from the fact that the option value has decreased but recall that when a bond is in the range of “negative convexity”, a rise in interest rates will cause the bond value to decrease MORE than if the bond was in the positive convexity area. In other words, the bond price decreases at an increasing rate.
No, when Intrest Rates are rising you dont want Treasuries, you want the spread products typically.
And what are considered spread products?
Your MBS, ABS, Corporates…anything that trades with a Spread over Treasuries.
bigwilly…what did you think of my post above regarding the MBS outperformance… This is the last piece for me to fully understand this. - If rates are currently low (MBS in the negative convexity area), and there’s a subsequent rise in interest rates, many people are saying that MBS will outperform Non-Callable Bonds because the value of the option drops mitigating some of the loss on the MBS security from the int. rate increase whereas Non-Callable bonds do not have that mitigating factor. HOWEVER, the definition of “negative convexity” states that as interest rates rise, prices will drop at an INCREASING rate relative to those bonds that are in the “Positive Convexity” area. So if interest rates are currently low and MBS is in the negative convexity area, while non-callable bonds are in the positive area, is the change in value option sufficiently large to make up for these differences in convexity and the impact on the bond prices?
I beleive MBS securities will Outperform more than likely and at teh very least = the return on Non-callables I believe.