PV01 AND IE01: MODELS ON MODELS

In the last six years or so, since Redington was founded, the terms PV01 and IE01 have become part of the language of pension scheme trustees and actuaries, when previously they were restricted to Fixed Income traders and structurers in investment banks. PV01 is the change in present value of an asset or liability for a 1 basis point change in the nominal yield curve used to value the asset or liability (usually the swap curve); IE01 is the change in present value of an asset or liability for a 1 basis point change in the implied inflation curve used to value the asset or liability (usually the RPI zero-coupon curve). The gradual prevalence of these two terms has undoubtedly signified a better understanding of the risks inherent in DB pension schemes.

Measurement of these two factors, though, presents difficulties. While calculating PV01 for a fixed/floating swap is a relatively straightforward calculation – any first year analyst on a trading desk or investment consultant quickly becomes familiar with it – IE01, in the context of a pension scheme, is complicated by the presence of caps and floors on inflation increases in the benefit structure. This means that an IE01 value can vary materially depending on who calculates it.

A traditional approach to calculating IE01 was simply to take the “binary” view that, if expected inflation was above the cap at a certain level, we would assume that cashflow behaved like a fixed cashflow. If it was below the cap, it behaved like a fully inflation-linked cashflow.

However, a little thought and capital markets knowledge quickly made people realise that various option-pricing models could be brought to bear on the problem and that, in fact, the true value of the sensitivity of the LPI benefit to changes in RPI (in the language of option pricing, the delta) may not be 0% or 100% but somewhere in between. These option pricing models could be readily calibrated to the market-observed levels for LPI swaps, which have been quoted by a number of banks since 2007.

Although using LPI swaps to calibrate models in order to calculate IE01 is logical, over time the price of LPI swaps has changed quite significantly (see chart). In some cases, this change in price has led to varying outputs from models which previously agreed very closely.
 

Graph-Dan-LPI05-v-RPI-Spread.png

If I gave the same fixed cashflow to four different fund managers I would expect them to agree on the PV01 of that swap to within less than 1%. The calculation is standard and, on the whole, unchangeable. However, if I were to give the same LPI cashflow to four fund managers, it is quite possible they would not agree on the IE01 to within 20% of each other.

The calculation of IE01 lacks a prescribed standard; values of the IE01 of your pension scheme could be materially different to those identified by another advisor or fund manager, and this could have an impact on the portfolio that would most closely hedge the liabilities. While the perfect formula for an IE01 calculation is still an elusive concept, pension funds should be asking their advisors who calculates this value for them, and asking for more clarity in order to initiate a debate on the correct approach.

Something so influential in the hedging of liabilities needs to be better understood, and pressure needs to build to find a more accurate and standardised measurement.
 

[Please note that all opinions expressed in this blog are the author’s own and do not constitute investment advice. Click here for full disclaimer]