Page 1 of 1

Did T-Bills Really Lose 20% in 2008?

Posted: Tue Mar 05, 2013 4:25 pm
by MachineGhost
9/18/2008 28.52662792
9/19/2008 22.83347847
9/22/2008 28.44137302

???

Re: Did T-Bills Really Lose 20% in 2008?

Posted: Tue Mar 05, 2013 4:31 pm
by MachineGhost
Nevermind, it seems a gremlin popped into my total return cash data.  Not sure how that happened.

Re: Did T-Bills Really Lose 20% in 2008?

Posted: Tue Mar 05, 2013 4:47 pm
by MachineGhost
I believe this one is legitimate.  Rates went from .50% to 1.35% overnight, resulting in a 15.66% principal loss.

3/27/2008 19.67741014
3/28/2008 16.59586231
3/31/2008 19.64891648

Anyone think otherwise?

Re: Did T-Bills Really Lose 20% in 2008?

Posted: Tue Mar 05, 2013 4:52 pm
by D1984
MachineGhost wrote: I believe this one is legitimate.  Rates went from .50% to 1.35% overnight, resulting in a 15.66% principal loss.

3/27/2008 19.67741014
3/28/2008 16.59586231
3/31/2008 19.64891648

Anyone think otherwise?
That makes no sense...for a six-month t-bill to lose nearly 16% means it has essentially become a very short term zero-coupon bond with an interest rate upwards of 30% assuming it will be redeemed at par in 6 months. That's far more a reaction than you'd expect for a 0.85% change in rates for an instrument with such a short maturity and duration.

Re: Did T-Bills Really Lose 20% in 2008?

Posted: Tue Mar 05, 2013 5:17 pm
by MachineGhost
D1984 wrote: That makes no sense...for a six-month t-bill to lose nearly 16% means it has essentially become a very short term zero-coupon bond with an interest rate upwards of 30% assuming it will be redeemed at par in 6 months. That's far more a reaction than you'd expect for a 0.85% change in rates for an instrument with such a short maturity and duration.
I agree and the problem must be in Excel's PRICE function, but I can't see anything wrong with it.  Maybe PRICE doesn't work for less than a year to maturity?  I am going to attach the Excel spreadsheet with the March 2008 T-Bill Yields, so if anyone wants to take a gander at fixing the total return, be my guest.

http://www.docyoushare.com/file/index.php?f=CzMRbeHq

Re: Did T-Bills Really Lose 20% in 2008?

Posted: Wed Mar 06, 2013 1:04 am
by Tortoise
I can't seem to access your Excel file, MG, but my hunch is that you're plugging in the coupon and yield as percents (i.e., 0.5 and 1.35) instead of the unitless fractions (i.e., 0.005 and 0.0135) that Excel's PRICE function expects.

Here is what I calculate in Excel assuming a 1-year T-bill with a coupon of 0.5% responding to a yield change from 0.5% to 1.35% overnight:

PRICE(DATE(2008,3,27), DATE(2009,3,27), 0.005, 0.005, 100, 2) = $100.00
PRICE(DATE(2008,3,28), DATE(2009,3,27), 0.005, 0.0135, 100, 2) = $99.16

Single-day change in price: -$0.84 = -0.84%

In other words, as we would expect, the 1-year T-bill single-day price drop in percent is approximately equal to the single day yield change in percent.

Re: Did T-Bills Really Lose 20% in 2008?

Posted: Wed Mar 06, 2013 9:16 pm
by MachineGhost
Tortoise wrote: I can't seem to access your Excel file, MG, but my hunch is that you're plugging in the coupon and yield as percents (i.e., 0.5 and 1.35) instead of the unitless fractions (i.e., 0.005 and 0.0135) that Excel's PRICE function expects.
You are money!  Thanks.