How the Pilkington Superannuation Scheme completed a £230m buy-in

Jonathan Stapleton: What were the key reasons for you undertaking a buy-in and what was the background to this deal?

Keith Greenfield: The scheme, like a lot of defined benefit (DB) schemes, is relatively mature - and closed to new members though not to future accrual. The employer, a glass manufacturing company now owned by a Japanese parent, Nippon Sheet Glass Company (NSG), was keen to work with the trustee to de-risk and minimise any further increases in the commitments it has made to the scheme. So unlike a lot of employers - who perhaps want to keep on risk while the trustee is pushing for de-risking - it was very much about the employer working with the trustee to bring about de-risking.

We'd already done quite a bit of inflation and interest rate hedging; and had also worked on the investment portfolio to get that into a relatively low risk position but with some growth capability.

We had additionally completed a £1bn longevity swap with Legal & General in 2011, which covered all the pensions in payment up to that date but excluded dependents.

However, since then, new pensioners and dependents' pensions had come into payment and longevity risk was increasing again.

And that is the background to the buy-in we conducted last year - we held the view that we had done quite a lot on the investment portfolio; we had reduced longevity risk and there was now was an opportunity to do a bit more.

Jonathan Stapleton: What group of the membership did the buy-in cover?

Keith Greenfield: The buy-in covered all the people whose pensions came into payment after the longevity swap until the point we completed the buy-in last year and also covered dependents.

Jonathan Stapleton: When did you start to think about conducting a buy-in and what steps did you go through to prepare?

Keith Greenfield: We had been thinking about a buy-in transaction since just before the 2015 General Election, when we decided to do some more preparatory work on our data in preparation for a potential deal and look at how many of our pensioner members were married and what was the age of their spouse.

People often don't do a great data gathering on this; they just assume 80% or 90% of them are married and that the spouse is three years younger than their partner. And they just work on those assumptions. Our view was if we could get some more accurate data on that, it might help with a quote.

We decided to write to all our pensioners asking them if they were married and, if they were, how old their spouse was. We got a huge response to this and over 90% of our pensioners responded - giving us some really accurate data.

When we compared these numbers with our valuation assumption, it wasn't that different - maybe an odd million or so worse than what we thought in terms of added liability - but we felt, and our advisers were telling us, that insurers should be able to give us a keener quote because they are not having to make an assumption in this area, they have got some real data to work on. So we felt that was a useful exercise to go through.

And then we came into 2016 and decided to transact. Obviously the EU referendum cropped up in the middle of it but, by this point, we had been through the tendering process - moving from a longlist to a shortlist and had chosen PIC as our supplier. But, as part of the final negotiations, we wanted to try and secure the price as much as we could, locking in the price relative to our gilt portfolio.

We decided to write to all our pensioners asking them if they were married and, if they were, how old their spouse was. We got a huge response to this and over 90% of our pensioners responded - giving us some really accurate data.

Uzma Nazir: When we were chosen as the provider, there was a period of a few weeks between being selected and contracts being negotiated and actually signed. So there is always a period when the price will change from an insurers' point of view. The trustee wants certainty as to how that will change and we often enter into arrangements where we predefine how our premium is going to move, and there are various metrics for doing that. What we agreed on Pilkington seemed to work on both sides; it worked for us and it worked for the scheme in terms of how it was measuring liabilities.

Jonathan Stapleton: Did Pilkington benefit from market volatility following the EU referendum?

Uzma Nazir: Pilkington was holding gilts for the transaction and our price generally moves in line with gilts and corporate bond spreads. Just after the EU referendum, corporate bond spreads rose quite a lot which meant we could buy corporate bond assets at attractive rates.

Our pricing reflected that and that's what the scheme ended up locking in to - as they are measuring liabilities against gilts and not corporate bonds, then that will be beneficial because our price will have gone down whereas their liabilities won't have changed because of corporate bond spreads. So, Pilkington benefited from that in our pricing and effectively locked in to this pricing just after the referendum.

Jonathan Stapleton: Is there any advice you would offer other schemes looking to conduct a similar deal?

Keith Greenfield: There are a number of things. One is to be well prepared and plan ahead. Don't think you can just do this overnight - it might only be a couple of months from start to finish when you're well prepared and deadly serious but, if you need to get your data in place and understand at what price you are willing to transact, then you have to start planning well in advance. So, be well prepared, plan ahead and sort your data as much as possible.

The second thing is within the trustee board itself. Most boards probably don't move as quickly as perhaps they could so what we did is make sure we had educated the whole board about what we were trying to do - holding a number of different training sessions about what buy-ins were, what buyouts were, what longevity swaps were, what we were doing and why we were doing it.

But then we also set up a much smaller committee, including representatives from the sponsor, that was given authority to move much more quickly. We delegated as much as we possibly could to that, obviously going back to the main board when we had final pricing and deals but we set up separate meetings for that. So having a committee that can move quickly is important.

I think going to tender is also important - it sounds obvious but it creates that competitiveness in the process. So we went from about six or seven in the first round and then we cut it down to two or three after that. I suppose what's surprising is there can be quite different movements between the different parties on the second round as they actually firm up their offers and actually decide if this is a piece of business they want or not. So we actually saw what was seemingly the best offer get usurped by PIC in that second round. I think it is important for the trustee to have that competitive tension in order to secure the best price.

You also need to have an idea at what level you're willing to do the buy-in and to know what level you feel is good value. Going in without that leaves you very exposed because you'll get a price and you have no idea whether that's good value or not and whether or not you are paying too much. So I think doing some preparation about the level at which you're willing to complete a deal is important.

Clearly the use of your main adviser is also important; we used Aon Hewitt, who are our scheme actuary, and we are very pleased with the work they did for us. But I think more important is how all the advisers then work together because you cannot work in isolation just on a buy-in, you have got to work on what's going to happen to your investment portfolio, what investments need to be ring fenced and ready to be passed over and transferred. And therefore it's important that all the advisers, including those acting for the sponsor, are working with each other to make this happen.

I think probably the most important thing is that you have got commitment from your scheme's sponsor, because without that, any transaction will fail. So many trustees go into this saying we want to do a deal and then turn back to the company, which says it is not willing to do it for whatever reason.

And there can be all sorts of different reasons - they may not think the pricing is that good; they may be willing to carry more longevity risk. One of the things is that company accounts are driven by IAS19 numbers. And when you do a buy-in, it can have an adverse effect on the IAS19 deficit or surplus because assumptions change. So this does have an impact on the sponsor and they've got to be willing to take a possible hit on their IAS19 numbers and be prepared to communicate to their shareholders and analysts about what's going on. Because the reality is, while IAS19 is just an accounting number, this is a real de-risking issue. So you do need the full support from your sponsor, otherwise it just won't happen.

Finally, I would say you need to be serious about doing the deal and not waste anyone's time here. Insurers get lots of people coming to them for quotes, but maybe some who just want to get a flavour of where the market is. And they have to spend a lot of time doing a quote and a tender and if they do loads and loads of those and nothing ever completes, then it probably becomes very frustrating from their point of view. So I think if everyone is serious about doing a deal, that's when things can happen very quickly and get done to the benefit of all parties.

Uzma Nazir: I would completely agree. You can really tell the seriousness of the proposition by the level of engagement you get from the advisers and the detail you get in the request for quotation (RFQ) that is sent. I think the best processes are those where it's clearly laid out who's involved in the transaction and whether it's got a joint working party or not, who's going to make the decisions, if the meetings have already been booked in and exactly what their pricing metrics are. And Pilkington had exactly that.

But, as Keith said, we often get quite speculative quotation requests come in where you submit the quote and then you don't hear anything for months and you are not really sure what's happening. And, as an insurer, if you don't know what's happening, you can't really push your board and investment committee to allocate the best assets and the best pricing to it because you just don't know what's happening in the process.

You can really tell the seriousness of the proposition by the level of engagement you get from the advisers and the detail you get in the request for quotation that is sent.

Jonathan Stapleton: What are your future plans for risk reduction?

Keith Greenfield: I doubt we will be doing a buyout in the near term, but we have agreed a longer-term plan of getting to 100% on a technical provisions basis, which is a reasonably strong basis. After that we will look to move to 100% on a valuation basis, which is almost gilts flat, and then look to move to a place that is a bit more than gilts flat. As we move through those three phases, we would expect the investment portfolio to continue to de-risk until the point we will be holding very low risk-matching assets alongside the buy-ins. And we will continue to pass more assets over to insurers as more pensioners come on - giving us the certainty of totally matched cash flows coming in to meet pension payments. So I think that is going to be our way of eventually moving this scheme to self-sufficiency, with very low risk reliance on the company.