Many Happy Returns! #LDI

Fifteen years ago today, we were part of a team at Merrill Lynch that implemented a transaction which would change the way pension funds approached the vexed question of managing risk: how to best invest a pension fund's assets to ensure that the members would, for the next 50 years, be paid their pensions in full and on time? Up until then, pension funds by and large engaged in a precarious balancing act between investing mainly in bonds, on the one hand, and equities on the other. Pensioner liabilities were classically backed with bonds, whilst the liabilities attributable to Deferreds and Actives relied upon the out-sized returns expected from Equities.

The thesis that drove this approach was simple enough: Pensioner liabilities are short term and pretty well defined, and so can be matched by fixed income bonds, which produce reliable cash-flows year after year. Deferred and Active members are younger and it's a much harder science to predict how long they will live or what their pensions will actually be. So, the reasoning went, if you back those guys with equities (which are hard to predict in the short term but reliably produce great returns over the long term) you've got a workable investment strategy. And, up to 2003, that's how it all worked. Kind of.

The problem was that it didn't work very well. For two reasons. First, against the received wisdom and expectation, equities failed to deliver the promised long term out-sized returns (aka the "Equity Risk Premium"); second, real yields plummeted and the present value of pension liabilities soared, creating gargantuan deficits on sponsors' corporate balance sheets that the accounting gods decreed had to be dealt with. There were several other reasons but those were the biggies. 

It's worth taking a moment to think about that elusive Equity Risk Premium. It was well recognized back then that its occurrence was difficult to explain. In fact, one scholarly article was entitled The Equity Premium: A Puzzle. Read it, if only to gain an appreciation for the impossibility of the task.

Here's a flavour from page 151(!):

Exactly.

Anyway, our approach was not to rely upon the Equity Risk Premium to save the day. Instead, we layered a derivative (an inflation linked zero-coupon swap) over the top of the physical assets and structured so that the pension fund (our client) was completely immunized against the volatility of the real yield (believe me, it was insanely volatile) but, crucially, could still invest in equities and other equity-like assets if it wanted to. It worked like a charm for the pension fund (Friends Provident Pension Fund) and a year later the FT wrote it up.

Today, this is basic stuff known, of course, as LDI and this approach to risk management is the fundamental bedrock of managing pension fund risk. In summary, you right-size the risks in the pension fund and LDI is the starting point. Over 2,100 LDI mandates now hedge around a trillion sterling of pension assets, representing about half of all UK defined benefit pension schemes.

So it seems only right to throwback to 11:00 on that cold Tuesday morning in December 2003 and recall the team that did that transaction. It seemed epic at the time. Looking back, it was. 

@philip rose @jonathan mitchell @robert gardner @wilson chan @david gu @mark duke @faisal rafi @paul cooper