LON:JPR | Pension Deficits - Has The Worm Turned?

8th Feb '16 by Ramridge

17 comments 18929 reads

On February 3rd, Johnston Press Johnston Press (LON:JPR) put out an RNS stating that following an independent review their pension deficit has been reduced from £90m on 3/1/15 to £53m on 3/1/16. Following the announcement, the share price jumped 12%.
This is a massive deficit reduction of 41% at one stroke which is unusual but not entirely unheard of. The company also stated that full details of the study and assumptions used in calculating the changes will be presented on 22 March 2016.
In this note, (1) I will look at some of the details behind the announcement, (2) consider any read-across and (3) provide a partial list of companies with current pension deficits.

Detailed Look
A pension deficit is the difference between two very large numbers, pension assets and pension liabilities. In the case of JPR, the deficit of £90m was the difference between £480m of assets and £570m of liabilities.
The actuarially calculated pension liability is based on a number of assumptions, the key ones being net discount rate, inflation rate, mortality and longevity rates.
Ahead of the full details to be disclosed on 22 March, it is clear that the company has assumed a higher discount rate and possibly changed the mortality and longevity assumptions. Out of all the liability assumptions, the most sensitive is undoubtedly the discount rate. In fact the company has been very helpful in their 2014 accounts by providing a sensitivity analysis.
A 0.1% increase in the discount rate would reduce the pension liabilities by £9m. The discount rate they used on 3/1/15 was 3.55%. A £37m reduction therefore equates to an increase of 0.41% making the new discount rate 3.96%. Of course the actual new discount rate would be less than this if mortality and longevity rates have also been revised favourably.
So what we can infer from this is that a small change in the discount rate, from 3.55% to say 4%, can make an enormous difference to the pension deficit. In the case of JPR, it reduced the deficit by some 40%.
The impact on the accounts is as follows. (1) The balance sheet is strengthened by reducing the long term liabilities by £37m. (2) It will allow the company to reduce their annual deficit reduction contributions (slated to be some £10m in 2016/17). Any reduction will need to…

Unlock with Email

Unlock with Google

Already have an account?
Login here

WarrantStar

8th Feb '16

Very interesting analysis. I agree that that this could be relevant to a lot of other companies

shipoffrogs

Does what you describe here change how much this company's pensioners need to be paid in the future? No.

Does the market movements this year reduce the assets available to meet those liabilities? Yes.

The outlook for triennial valuations being struck this year looks pretty grim right now and will continue doing so whilst we live in a low interest environment.

pooledaniel

9th Feb '16

I would advise a bit of caution around assuming that other company's with pension deficits will benefit from a similar reduction. There are a huge number of factors at play here and though you have touched on some of the most important, there are lots of other details to consider. I'd be keen to take a look over the full report when it comes out as it will give rationale for change. In general though, things probably do look slightly better now than they did a year or two ago. Some comments below.

The implied term of the pension scheme liabilities is c.17 years (backed out from the discount rate sensitivity, assuming 0.1% increase in DR reduces liabilities by 9m, where liabilities start at 570m). Looking at the period from 3/1/15 to 3/1/16, the 17 year spot gilts (which drive the discount rate) have gone from 2.32% -> 2.53%. 17 year spot inflation has gone 3.28% -> 3.32%. This suggests an approximate improvement in net rate of 0.17%. This will account for a portion of the change. Worth noting that this will represent a general improvement in funding positions for all schemes that are not fully hedged (assuming no change in the methodology used to derive the DR / Inflation related assumptions).

The rest of the change may be a combination of changes in demographic assumptions. Perhaps surprisingly, the most recent estimates of future improvements in mortality are weaker than those published a few years ago. This means most schemes will get a modest reduction in liabilities from updating for this when they have their next triennial valuation. It may be that the other mortality assumptions (i.e. base tables, scaling factors, long term trend for future improvements) get updated to something more prudent, which could offset the above.

Over the past couple of years, inflation has been low. Where scheme's use a flat inflation rate (i.e. not a curve which allows for lower infl. in the early years), most schemes will be seeing an 'experience gain' where actual inflation was lower than assumed. As an example, in simple terms; they thought pensions would increase in 2016 by 3%, but in reality they will only increase by 0.8% (based on September 2015 RPI, which is the typical driver for pension increases and revaluation). This gives a 'saving' of 2.2% for all pensions linked in this way. Obviously there is a lot more complexity than this (different inflation linked increases, fixed increases etc), but you get the broad idea.

On the asset side - this is completely scheme specific. Never forget that deficit is a function of liabilities AND assets. If assets are very well hedged vs. inflation and interest, then the improvement in net rate set out above won't really be a benefit to a scheme. This obviously works in reverse if yields were to worsen. JPR appear to be running a 50% growth, 50% matching strategy which is generally quite risky. This may benefit, or may cause problems - almost impossible to say. Worth noting if the discount rate is an asset-led discount rate, then it may be a little higher than a typical DR.

I realise I'm rambling a bit and this isn't super coherent. In short though, there's a lot at play and no two schemes are the same, so take care with the thought process when applying this to other co's with big pension deficits. I guess I was just trying to highlight some areas that sprang to mind. Very happy to answer any specific questions you may have.

(I'm a pensions actuary, so spend my life playing with assumptions to get to the right figures... hence I'll always be sceptical around any 'pension deficit improvements')

Ramridge

pooledaniel -
I think I have been careful in saying that we need to wait till the full details are published and that there may well be JPR specific factors at play here.
Anyway, from a retired actuary to a pensions actuary, let's just agree that there are many moving parts at play and I have just highlighted some of the key ones.

Apologies if I came across as negative or condescending - didn't mean to at all. I agree that it's something worth looking at, just always tend to be cautious as it's generally very hard to assess pension schemes properly from publicly available information.

You've got me interested in this one now anyway!

No offence taken. The central question I asked myself was: has JPR found a way to reduce its pension deficit that can be applied by others who are equally pension-deficit-challenged? Because if they have, then sure as eggs are eggs, they will be queuing up and knocking on your door, as a pensions actuary, to help them boost their balance sheet, profits and eps. The impact on JPR's share price was remarkable.
Hence the blog.

Mark Simpson

Normally I prefer to look at pension deficits by adapting my valuation to exclude the recovery payments from the companies' profit figures rather than take off a deficit balance sheet figure from a unadjusted valuation. However something that made me more cautious about pension deficits was this announcement by Ensor Holdings (LON:ESR):

http://www.investegate.co.uk/ensor-holdings-plc--e...

Our balance sheet carries a gross liability of £2.0m in respect of retirement benefit obligations under the Ensor Group Pension Fund. This liability is currently financed by a schedule of contributions agreed with the scheme trustee, and paid by the Company to the scheme. Given the expected outcome of the Group sale process, we have now decided, in principle, to purchase an annuity which will secure all future liabilities of the Ensor Group Pension Fund, as a precursor to a buyout and wind-up of the scheme. The cost of this exercise, which will be payable in cash, is expected to be in the region of £5.5m to £6m and will be financed through short-term borrowings, to bridge the shortfall until further asset sales are realised.

I've not looked into the detailed assumptions behind the Ensor Holdings (LON:ESR) scheme and how they compare to Johnston Press (LON:JPR) but the fact that a £2m balance sheet liability cost c£6m (I confirmed with the company that this was in addition to the schemes assets so a true comparison) to exit suggests that caution is still warranted in this area. The difference was presumably due to the difference between the discount rate on the scheme valuation (where the company is taking the risk) and annuity rates (where someone else is taking the risk.) This highlights that there still is risk present in companies with large deficits since if they were to pay someone else to take that risk they would in most cases pay a lot more than the liability they have on their balance sheets.

Danger - might be worth a quick read of my comment in here: http://www.stockopedia.com/content/norcros-pension-problems-112809/?comment=16#16

Similar thing happened there re: Buyout deficit compared to accounting position. It really feels like pensions obligations are not communicated well in accounts, which leads a lot of people to misjudge the costs / risks associated!

Edit: Just realised you already commented over there - ignore my post!

Good to be reminded of the proper technical terms behind my thinking though, thanks!

10th Feb '16

very welcome. I thought it might be interesting to show the differences on a few schemes I work on (anonymous of course!)

Example 1:
Accounting liabilities = 460m
Scheme funding liabilities = 605m
Buyout liabilities = 730m

Example 2:
Accounting liabilities = 780m
Scheme funding liabilities = 900m
Buyout liabilities = 1,060m

Example 3:
Accounting liabilities = 370m
Scheme funding liabilities = 440m
Buyout liabilities = 550m

Thanks for the examples. Looks like an average of 22% increase of Scheme Funding liabilities compared to accounting liabilities and +48% for Buyout.

As an example if we look at Airea (LON:AIEA) . This reports an Accounting liability of £7.4m which is already large compared the it's £8.5m market cap. But when we look at the details in the FY15 accounts:

http://www.aireaplc.com/download-attachment/776/

we see scheme liabilities of £51.9m and scheme assets of £44.5m. Applying the +22% to estimate Scheme funding liabilities the deficit would be £18.8m. This dwarfs the market cap & c.£1m operating profit.

I'm quite surprised that the scheme's trustees have allowed the company to make just £400k overpayment this year and allowed them to pay a dividend. I think this is one case where my usual method of adjusting the op profit for the overpayment is not a good way of valuing the company.

The counterpoint of course is that I thought the pension deficit made the company completely uninvestable when Airea (LON:AIEA) traded at 10p and it hasn't stopped it rising to 22p today. It may be a greater fool theory to buy something with such large hidden liabilities hoping others will not care too, but so far the fools are winning!

AnonymousUser192020

I think it's dangerous to extrapolate from accounting and buy-out liabilities, which are both pretty tightly defined, to scheme funding liabilities, particularly from a sample of just three! Funding liabilities show appreciably wider latitude - that's why the regime we have is called the "scheme specific funding regime". In particular, strength of the employer covenant has a crucial impact on the trustees' funding basis but is largely irrelevant to the other two funding measures.

This article from yesterday is interesting http://www.professionalpensions.com/professional-pensions/news/2446314/ppf-7800-deficits-swell-27-in-january The full PPF 31 January 2016 report is here, complete with several charts http://www.pensionprotectionfund.org.uk/DocumentLibrary/Documents/PPF_7800_February_16.pdf Bear in mind that the PPF section 179 liability (yet another liability to complicate matters!) is not the IAS19 accounting liability. One major difference is that discount rates are referenced by gilt yields rather than AA bond yields (which typically were around 0.75% higher, though I haven't checked recently). This increases the liability figure. Offsetting this, several reductions are made to the liabilities to reflect how member benefits are cut back on entry to the PPF. That said, trends in PPF liabilities can be a useful pointer, if allowance is made for changes in the AA yield premium.

As others have said, the other big impact on a scheme's funding position is the degree of investment hedging, specifically equity risk relative to a hedged bond/ gilt portfolio. This will vary considerably between schemes.

All in all, it's difficult to extrapolate from the one example of Johnson Press. Investment analysts will of course apply their own updating adjustments when pension liabilities are large in relation to market cap.

Pooledaniel, it's a good job the accountants give you just one number for the assets.

11th Feb '16

Hi millen -
Thanks for the two links. I hadn't seen these before and found them very interesting.
I concur with your view and pooledaniel's that any kind of extrapolation from the Johnston Press announcement is pure speculation. We definitely need to wait until they publish the details in March, and even then I suspect a number of questions will remain unanswered.
Regards, Ram

I see MKS this morning announced an improved scheme funding position at 31 March 2015. There's no mention of deficit contributions - presumably there are none, though the position may well have worsened since the valuation date given the like fall in return-seeking assets since the equity market highs.

12th Feb '16

Fully agree with your post. Those examples were it no way intended to be used to try and proxy other companies SF liabilities, more just a demonstration of the variability between bases. The only real certain is that Buyout > Scheme Funding > Accounting. Where the upper and lower ends sit, and where SF sits within that is completely scheme specific. I think it's safe to assume that buyout will always be substantially higher than accounting (as by definition accounting is ~best estimate, where buyout is incredibly prudent).

Employer covenant point is a good one. This has a huge impact on the basis used for scheme funding, as well as the investment strategy and the way any recovery plan is set up (i.e. can have a big impact on timing of contributions etc).

PPF isn't a bad proxy really, thought I'm not sure whether the link you put up is talking about S179 liabilities based on assumptions as specific dates, or the smoothed/stressed approach that the PPF uses when rolling forward S179 liabilities to calculate underfunding for levy purposes. If it's the latter, then the valuation is based on assumptions derived using average yields over a 5 year period rather than a specific date, so you get some distortion where swings up or down are masked.

May be of interest generally: http://www.investmentweek.co.uk/investment-week/news/2446870/gilt-yields-fall-to-record-low-as-investors-flock-to-safe-havens?utm_medium=email&utm_campaign=IW.Daily_RL.EU.A.U&utm_source=IW.DCM.Editors_Updates

In short - gilt yields still falling, and equities falling too. Schemes which aren't well hedged against interest or have high equity exposures (which often come together as one!) will be having a very bad time at the moment. 20 year gilts have also fallen by 40bps already this year... not good news.

Edit: Millen - do you work in the industry by any chance? Or have done in the past?

23rd Mar '16

JPR's full year results were published yesterday. Here is an update to my blog.

In Feb the company had intimated a pension deficit reduction from £90m to £53m. They have actually achieved a substantially higher reduction. The new deficit is down to £27m.

Almost all of the reduction has been achieved by assuming a reduction in assumed life expectancy. As far as I can see the financial assumptions have remained unchanged.

So what is the read-across for others in a similar situation? If the KPMG study was able to justify the changes in demographic assumptions by a combination of a review of the members profile and use of new actuarial tables, then I can see the door being opened for others to follow. Whether they can achieved similar reductions will depend on the scheme specific members & pensioners demographic profile. But the temptation is there for the pension-challenged plc s to follow.

Pension Deficits - Has The Worm Turned?

Unlock the rest of this article with a 14 day trial

About the Author

Ramridge

17 comments

Unlock the metrics for all these stocks and more...

Unlock the metrics for all these stocks and more...