Liability Driven Investment

6. Alternative approaches

[this page | pdf | references | back links | custom searches]

Return to Overview

Next page


6.1          The overlay approach described in Section 5 clearly demarcates who is responsible for what. But trustees might prefer merely to set their investment manager a liability driven benchmark akin to the one described above, and say “get on with it”, with the investment manager free to use whatever instruments it likes (including swaps and other derivatives) and whenever it likes, to match the liabilities or preferably to add value versus them.


6.2          Key requirements for such an approach are for the trustees and their consultants to carefully craft an appropriate liability driven benchmark as above, for the fund manager to have good systems for measuring at all times how far its portfolio deviates from this benchmark and for it to be very clear exactly what is expected of the fund manager. The bespoke nature of such a service is likely to make it practical only for larger accounts. It is worth noting that if the fund manager cannot practically hedge a particular part of the liability benchmark then there will be a ‘random’ element to his performance. The fund manager may stress this whenever he thinks it has worked to his disadvantage, and the trustees may do the opposite whenever they think it has worked in the fund manager’s favour. Unfortunately, there is almost certain to be disagreement about which is the case, unless the whole arrangement is very carefully managed. An advantage of the swaps overlay approach described above is that it airs and manages these potential disagreements at outset, via the discussions needed around the formulation of the swaps overlay.


6.3          The trustees may deliberately want to adopt a strategy that deviates from the most precise liability driven benchmark. In these circumstances, a clear liability driven benchmark might still be defined but then deliberately modified to focus on what the trustees want.


6.4          For example, the trustees may feel that banks might be quoting excessive prices for buying cash flows that embed option-like inflation characteristics such as those implicit in LPI linked benefits. Yet they may still want some hedging of such risks. They might then ask the fund manager to hedge these risks in a more approximate way, using dynamic hedging, to avoid ceding this supposed profit margin to the bank. This could perhaps most easily be achieved by giving the investment manager a benchmark that changes in a dynamic fashion as the underlying economic parameters change. The aim would be to mimic the economic sensitivity of the fair value of the option-like characteristics insofar as far as these depend on the parameters in question. A perfect hedging algorithm, were one to exist, would of course also depend on volatility, which would require the use of more complicated derivatives (but this would then defeat the point of seeking to avoid the use of such derivatives because they are believed to offer poor value-for-money).


6.5          Some modification to the swaps overlay approach may be needed for smaller schemes. A single swap might be easier to have ‘segregated’ in this context than a whole bond portfolio, but there are still implicit lower limits on the sizes at which they become practical. A better alternative may be to create specially tailored long duration pooled bond funds. Several investment managers appear to be designing such products. In real life, a portfolio of pension liabilities typically gets shorter over time, so any pooled approach is unlikely to match any particular scheme’s liabilities as well as a more bespoke approach.


Contents | Prev | Next

Desktop view | Switch to Mobile