Skip to content
All library documents

Brazilian Break-Even Inflation Curves by Bond Index

Article Quant Q&A · Author: pyCthon
Read the original

Summary

The document asks how to construct historical break-even inflation curves in Brazil, where two inflation-linked government bond series reference different price indices: NTN-C is tied to IGP-M and NTN-B to IPCA. The response recalls that Deutsche Bank used separate curves for the two index families, rather than combining them. This reflects that the bonds represent different inflation measures, so pooling them would blur the index basis represented by each curve.

The answer also cautions that break-even inflation rates may be weak predictors of realized future inflation in this market. It points to Brazil’s relatively short indexation lag, compared with US TIPS, and to illiquidity at some inflation swap maturities. Those features can impair interpretation and require adjustments to standard inflation swap pricing methods. The note is brief and cites recollection rather than documented methodology or quantitative evidence, so it gives market-practice context but not a full curve-construction recipe.

Key ideas

  • NTN-C and NTN-B reference different inflation indices, IGP-M and IPCA respectively.
  • The cited practice is to build separate break-even inflation curves for the two index bases.
  • Brazil’s short bond indexation lag may weaken the predictive value of break-even rates.
  • Illiquid inflation swap maturities can require adjustments to standard pricing methods.

Tags

Fixed incomeCommodities
Full text
# Brazilian break even inflation curves


# Brazilian break even inflation curves












Brazil has had historically two sets of inflation bonds, the NTN-B and NTN-C series each with a different inflation index. The NTN-C is no longer issued.

When creating historical break even inflation (BEI) curves for Brazil what is the standard practice? Do we create two separate curves or combine both series some how?

## Answer by kumarz (score 1)

https://quant.stackexchange.com/a/31914

yes: from what I can recall, Deutsche use separate curves for IGP-M(NTN-C) vs IPCA(NTN-B). Note also that the BEIR doesn’t predict future inflation very well, one of the reasons being that unlike other markets, the indexation lag of Brazilian real bonds is very small (only a half month, the lag for TIPS is 3mths), and that certain points on the inflation swap curve are very very illiquid. Certain tweaks need to be made to the usual inflation swap pricing methodology.

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.