John Harrison: Liability driven investment has arrived

It has always been difficult to reconcile the two central objectives applied to LGPS funds: to minimise contribution rates and to keep contribution rates stable. Minimising contribution rates is best achieved by generating high long-term investment returns from volatile growth assets. Keeping contribution rates stable is best achieved by investing in assets that are closely aligned to liabilities, such as long-dated index-linked gilts, but these assets offer prospective returns that are insufficient to keep contribution rates low. There is a fundamental conflict.

Historically, most LGPS funds have solved the dilemma by focusing on returns generation and smoothing out contribution rate volatility through tweaked actuarial assumptions, stabilisation arrangements and long deficit recovery periods. This seemed reasonable given the long-term nature of LGPS liabilities and in an era of strong positive cash flow. Unfortunately, the scope to continue with this approach is now limited.

Perhaps the biggest problem has been the persistent fall in real yields on long-dated index-linked gilts. Current actuarial methodology uses long-term real yields to derive the current value of liabilities. With real yields having fallen to well below zero, liability values have risen far more rapidly than asset values, resulting in significant funding deficits across the sector.

Concurrently, the ability of LGPS funds to smooth away contribution rate volatility has eroded. Outsourcing and academies have increased the number of employers, a growing proportion of which do not have the time horizon or risk tolerance to ride out the bumps. Even those that do are constrained by already assuming high prospective returns relative to liabilities and long recovery periods, so the scope to increase these further is limited.

The end result is that LGPS funds will find it increasingly difficult to ignore liability management as a way of making contribution rates more stable. Liability driven investment (LDI) seems set to become more widespread in the sector. How then might LGPS funds approach LDI?

We start from the advantage of having the trail blazed for us by corporate pension funds. LDI has been a prominent feature of corporate schemes for over a decade, with material evolution of techniques and instruments along the way. However, it is important to remember that the needs of the LGPS and the corporate sector are very different.

Most corporate schemes are closed to new members and often to future accrual, so the liability streams are more mature and more easily defined. They also typically have shorter time horizons, particularly if they are on a flight path to buy-out, and may have sponsors that can make significant one-off contributions to accelerate de-risking strategies. The end result is that many corporate schemes have quite complex LDI strategies that align precisely to their specific liability needs.

By contrast, LGPS funds are not seeking a buy-out exit and as open schemes they face greater uncertainty in their liability profiles. There is therefore much less need to match liabilities precisely. A significant reduction in contribution rate volatility can be achieved with quite simple LDI structures.

An LGPS fund that is considering adopting LDI should focus on the following key questions:

• Is LDI to be applied to the whole fund or as an option for individual employers?
  If at the individual employer level it will be necessary to offer a limited range of pooled LDI funds.
• What dilution in long-term returns will the board/committee accept?
  It is possible to limit the physical capital required for risk mitigation using leverage or replacing some passive equity exposure with derivatives, but it is unlikely that most LGPS funds could hedge much more than a half of their liabilities without allocating over 20% to LDI.
• What triggers will we set for making LDI investments?
  Most funds could not afford to hedge out all their long-term liabilities at current real yields. The heavy lifting of making de-risking affordable must therefore be provided by market movements – either at the overall fund level (an improved funding level), or specifically in the liability values (higher real yields on long index-linked gilts).

In the past, the LGPS sector has largely ignored LDI as either unnecessary or too expensive. Changes in the structure of the LGPS will make it harder to do so in future. Unfortunately, it is now more expensive than ever. Many funds will be reluctant to implement a meaningful LDI strategy when real yields are so low. However, there is still merit in establishing an LDI strategy and defining the trigger levels at which level implementation would be tolerable.

John Harrison is a director, multi asset at Henderson Global Investors and independent investment adviser to Surrey ad Northamptonshire pension funds.