By Alex White, Head of ALM Research at Redington
Corporate pension schemes are a set of promises from a corporate sponsor to pay pensions (other schemes, such as LGPS schemes, are in a fundamentally different position given the nature of their sponsors, so tend to use different solutions). A pension, at heart, is a stream of income. A typical DB pension is an amount of money known in advance that increases with inflation each year – they look a lot like bonds. The scheme has a pot of money to pay these pensions – failing this, it must rely on contributions from the sponsoring company. This is always a burden on the company to some degree, and in some cases, a crushing one (e.g. Kodak). So schemes try to invest to ensure they can pay pensions without undue reliance on the sponsor.
Critically, pension schemes aren’t in the business of buying the highest-returning assets for their own sake. They are investing towards a clear goal, and trying to make achieving that goal as likely as possible.
What if pensions paid gold?
Suppose a scheme didn’t pay a pension, but instead paid a block of gold on retirement. They would have different problems. There are plenty of portfolios the scheme could hold which could work. For example, equities have generally outperformed gold, so this scheme could hold equities and aim to sell them and buy gold as members retire. However, they would clearly run a lot of risk against the price of gold spiking, which often happens when equities fall. So in an equity crash, the scheme could suddenly find itself in trouble.
If this scheme had enough money, it could just buy gold (or gold futures) and use it to pay members. That would be ideal - but may not be affordable. If a scheme could afford to buy 99% of the gold it needed, it might hold plenty of gold and a small amount in equities to make up the deficit. We might expect the scheme to hold more gold as the funding position improves.
Even if the scheme needed a lot more money, we might expect them to take some exposure to the gold price. The biggest risk for this scheme would be gold becoming more expensive, so buying gold futures would reduce the risk and make it more likely that the growth assets could earn enough – in a sense, it would be fixing the goalposts.
Now suppose the price of gold fell 50%. This scheme would lose 50% of the value of its assets. That sounds terrible – but it would also only need half as much to pay its members. So it would be fine. For this scheme, it would be best to use the asset value divided by the price of gold required to pay members (the funding level) as a barometer of health rather than the asset value alone.
Linking it back
If your liabilities are based on gold prices, LDI would advocate you buy gold. DB scheme liabilities are income streams, so LDI implies buying income streams – i.e. bonds. ‘Locking in’ the price of income streams narrows the range of likely outcomes and generally makes it more likely that a scheme can meet its pension payments – and that, if it misses them, it misses them by less.
From a different angle
Reversing the problem, we can also ask how difficult it is to pay pensions. If gilt rates are 4%, earning 4% is easy – you buy gilts. If gilt rates are 0%, earning 4% is hard. Fundamentally, you need less money to pay the same pensions with the same degree of certainty when rates are high than when they’re low. So, in a world in which you want a high degree of certainty for the benefit of members, it makes sense for pensions to be valued against interest rates.
In conclusion
There are legitimate questions to ask about leverage and collateral management within LDI. The industry has roughly doubled the safety margins typically held, but there are always important details to clarify. That does not mean abandoning the whole edifice of LDI, which exists for very real reasons unrelated to ‘artificial’ accounting.
As a final point on policy, pension schemes are material buyers of government bonds. Rewriting legislation to discourage them from holding gilts could cause a far larger and faster sell-off, and a greater systemic shock, than the crisis experienced in late 2022.
* https://redington.co.uk/our-thoughts-on-the-house-of-lords-ldi-letter/
|
