Why Interest Rate Swap Risks Are Grouped into Tenor Buckets
Summary
The document explains why interest rate swap sensitivities may be assigned to a standard tenor bucket, such as reporting a two-month swap under a three-month tenor. Bucketing groups nearby maturities whose rates are expected to move similarly, making risk summaries and limits easier to use. The answer notes that firms can choose different internal groupings, while regulatory reports may require specified standard tenors.
The discussion distinguishes convenient bucket reporting from a fuller view of curve risk. It recommends also tracking sensitivities to historical principal components, and explains that P&L attribution should use the instruments that fit the interest rate curve. Reporting sensitivities can then be mapped from those fitting instruments to standardized tenors. The examples refer to regulatory tenor lists and linear interpolation as one assignment method. Buckets are approximations: their usefulness depends on how closely rates co-move and on the reporting purpose, so standardized reports need not replace finer internal monitoring.
Key ideas
- Tenor buckets group rate sensitivities at maturities considered similar in their market movements.
- Different firms can use distinct internal bucket schemes, while regulatory reporting may prescribe standard tenors.
- Principal component sensitivities can complement individual tenor buckets when monitoring curve risk.
- P&L attribution should use sensitivities to the instruments used to fit the interest rate curve.
- Standardized tenor sensitivities can be estimated from fitting instrument sensitivities for reporting.
Tags
Full text
# Tenor bucketing for swap interest rates? # Tenor bucketing for swap interest rates? in the place I work I've noticed that for asset class Interest Rate Swaps, tenor bucketing takes place. Example as follow: - IRS with maturity 2 month being bucketed into a "3 month tenor bucket" Page 32878 of Federal Register speaks about it as well: https://books.google.com.ar/books?id=vJ9D8jDiXjoC&pg=PA32879&lpg=PA32879&dq=bucketing+tenors&source=bl&ots=nSlabnT3pJ&sig=ACfU3U2vqAsc9K9MebJlIf-i-pDK62AI6g&hl=es-419&sa=X&ved=2ahUKEwifvYHX28_nAhUhFLkGHXkxCMcQ6AEwAnoECAYQAQ#v=onepage&q=bucketing%20tenors&f=false What is a logical reasoning to think this is acceptable? ## Answer by David Duarte (score 2) https://quant.stackexchange.com/a/51149 One possible reasoning is to group the underlying risk into similar categories. You could have 3m, 2y, 5y, 10y, 30y,... No written rule here. Different banks or traders may like to group the tenors in different ways. For example, the 4y and 5y swap will most likely always move very closely so you can group them together and look at your risk by buckets. Check out this post for an example of calculating tenor wise DV01. After having your portfolio risk separated by tenor you can group them to have a better view of your risk. ## Answer by Dimitri Vulis (score 0) https://quant.stackexchange.com/a/82408 The assumption is that tenors other than CFTC's (rather short) list of standard tenors are highly correlated to the tenors already on their list. Interest rate 01 (as well as credit spread 01) by standardized tenor bucket is convenient for reporting market risk and setting market risk limits. However it is a good practice, in addition to individual tenor buckets, to report sensitivities to (at least the first three) historical principal components of the interest rate curves, with bumps expressed in terms of their historical standard deviations. The cited CFTC document in the Federal Register can be found here: https://www.federalregister.gov/d/2013-12133/p-336 Here is another list of tenors buckets (Basel GIRR) https://www.bis.org/basel_framework/chapter/MAR/21.htm?inforce=20230101&published=20240705#paragraph_MAR_21_20230101_21_8 > 0.25 years, 0.5 years, 1 year, 2 years, 3 years, 5 years, 10 years, 15 years, 20 years and 30 years Footnote 3: > The assignment of risk factors to the specified tenors should be performed by linear interpolation or a method that is most consistent with the pricing functions used by the independent risk control function of a bank to report market risks or P&L to senior management. Two important points: - For the purposes of P&L attribution, you should use (all) the fitting instruments of your interest rate curve. So, for example, if you build your interest rate curve from futures, rather than swap rates, for tenors < 10 years, then you should calculate the sensitivities to the futures, by bumping the futures quotes and repricing; and use the sensitivities to the futures for P&L attribution; but use something like inverse Jacobian to estimate the sensitivities to "reporting" tenors from the sensitivities to the fitting instruments. - (Sane) regulators don't prohibit you from using additional tenor buckets internally, e.g. for market risk reporting and limit setting. If you want to monitor/limit your book's sensitivities, for example, to 1 month, 2 months, 9 months, 18 months, or 7 years - you are very welcome! But for some regulatory reporting purposes, you may need to rebucket into some standard buckets.
Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.