No. of Recommendations: 0
Loki,

Warning: Long, boring post. Conclusion at bottom.

Here's what 5-year TIPS have traded at over 6 months (October not listed),

Thanks for that link. I had previously looked at the Fred II database, and the closest thing I could find was the monthly data for a 10-year TIPS that was approximately 5 years from maturity:
http://research.stlouisfed.org/fred2/data/TP10J09.txt
or
http://research.stlouisfed.org/fred2/data/TP10J10.txt

I suppose the Treasury uses a blend of these 2 issues to determine the values in the constant maturity series you provided. Since that data begins with Jan 2003, we can look back only as far as the I Bonds issued in Nov 2003. :-( Here's what I calculated:

Nov 2003
If the period used was 05/03 - 10/03, the average yield was 1.23%. If the period was 04/03 - 09/03, the average yield was 1.25%. Too close to distinguish which period might be correct. 90% of either value is 1.1% (rounded to the nearest tenth). The fixed rate of I Bonds issued in Nov 2003 was ... 1.1%.

May 2004
11/03 - 04/04: 0.998%. 10/03 - 03/04: 1.03%. Unfortunately, taking 90% of these values ruins things. 90% of the former period gives 0.90%, the latter gives 0.93%. Neither equals the fixed rate of May 2004 bonds (1.0%). So this method seems close, but not entirely right.

Or maybe they are simply using the EE rate and subtracting 90% of how much on the average 5-year T bills have traded above 5-year TIPS

An interesting and intelligent idea. Looking back to Nov 2003, 90% of the difference between the average monthly 5-year TIPS and 5-year T Bills rates were 1.50% and 1.44% (again, depending on the exact period used). The EE rate for Nov 2003 was 2.61%. That gives hypothetical rates of 1.1 and 1.2 (rounding to the nearest tenth). Remember the answer we want is 1.1, so the former value is again promising. Let's look at May 2004:

90% of the difference gives 1.94% and 1.88%. The EE rate for May 2004 was 2.84%. Rounding to the nearest tenth, that gives 0.9% and 1.0%. Unfortunately, it's the second period (Oct-Mar) that's correct, whereas the former period (Apr-Oct) was the correct one for Nov 2003.

Which means that, unless my calculations are wrong (ALWAYS a distinct possibility), your last hypothesis is the only one I can't disprove:

Or maybe they throw pieces of paper down the stairs

Conclusion:
Both of your suggestions are very good ones, and they both come close to explaining the established rates -- but, unfortunately, not exactly right.

I wonder if anyone has ever simply asked the Treasury how they set the rate? I know I haven't.

Ken
Print the post  

Announcements

When Life Gives You Lemons
We all have had hardships and made poor decisions. The important thing is how we respond and grow. Read the story of a Fool who started from nothing, and looks to gain everything.
Contact Us
Contact Customer Service and other Fool departments here.
Work for Fools?
Winner of the Washingtonian great places to work, and Glassdoor #1 Company to Work For 2015! Have access to all of TMF's online and email products for FREE, and be paid for your contributions to TMF! Click the link and start your Fool career.
Advertisement