Why Pension Liabilities Can Respond Nonlinearly to Interest Rates

Summary

The document uses a simplified pension liability example to show why a liability's present value can respond nonlinearly to changes in assumptions. With a perpetual stream of payments growing at rate G and discounted at rate K, the stated present value depends on the spread between those rates. Its sensitivity to either input changes with that spread, rather than remaining constant, and can grow sharply as the spread narrows.

The practical implication is that a hedge built for current liability assumptions can become mismatched when those assumptions change. The example is deliberately stylized and assumes payments continue forever, so it is not a realistic valuation model for an actual pension plan. It illustrates the source of nonlinear exposure and hedge mismatch, but does not quantify a real fund's liabilities or prescribe a specific derivative hedge.

Key ideas

• A growing stream of liabilities can have present value that depends nonlinearly on its discount and growth rates.
• The sensitivity of liability value changes as the spread between discount rate and growth rate changes.
• A hedge calibrated to current assumptions can diverge from liabilities when those assumptions move.
• The perpetual-payment setup is an extreme illustration, not a practical pension valuation model.

Tags

# Hedging guaranteed liabilities for a pension fund


# Hedging guaranteed liabilities for a pension fund












I'm trying to understand what we mean by linear and non-linear guaranteed liabilities for pension fund ?

Often this exposures are hedged by Interest rates derivatives swaps and swaptions which I can understand just looking from the payoffs of those derivatives but what I can't still understand how liability can be non linear ? Thanks

## Answer by demully (score 1)

https://quant.stackexchange.com/a/57392

Consider a very simplistic example.

Imagine a fund with liabilities for this year of 1. These grew by say 3% per annum in perpetuity (call this "G"). The same discount rate applies to all liabilities, say 5% (call this "K").

So the Net Present Value of the future liabilities will be (1+K)/(K-G).

While the derivative of the NPV with respect to either K or G becomes a reciprocal quadratic of the spread between them... which is an explosive function, not a constant!

Please pay no heed to my implicit assumption about the pensioners being immortal... Above was a clearly extreme and stylised scenario; but hopefully reveals the underlying essence of the problem.

Which isn't the rates hedge the fund has on for today's future liabilities; but the mismatch that develops between this hedge and the liabilities once any of the assumptions used to calculate the latter get tweaked a little in any direction.

Maybe this helps?

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.