This abstract relates to the following paper: Charmaille, J-P., Clarke, M.G., Harding, J., Hildebrand, C., McKinlay, I.W., Rice, S.R. and Reynolds, P. Financial Management of the UK Pension Protection Fund. British Actuarial Journal, doi: 10.1017/S135732171200044X
The Chair (Mrs K. A. Morgan, F.I.A.): Before we start the meeting, Lady Barbara Judge, chairman of the Pensions Protection Fund (PPF), would like to say a few words.
Lady Barbara Judge: Thank you very much, Mrs Morgan, for giving us a chance to hear about this interesting topic and for organising this meeting in such an effective way. I just want to say how proud I am to be the chairman of the PPF. When I joined the PPF people were talking about it as a lifeboat that would probably sink. However, as we know, it is a lifeboat that is steaming ahead. I am proud of our CEO, Alan Rubenstein, and our whole team. The Government is now even asking us to help it with various problems that it is worrying about.
Today we are about to discuss this paper on a new idea: a case study on enterprise risk management. When I discussed this with our team, I asked “Why are we doing this?” They said, “Because it is about transparency; it is about telling the world how we are running the PPF.” I believe this is an interesting and important initiative and I would like to recognise the group that worked on it. Martin Clarke led the effort and Stephen Rice is going to close the meeting tonight. Ian McKinlay, Catherine Hildebrand, Jane Harding, Jean-Pierre Charmaille and Paul Reynolds all each also played an important part. On behalf of the Board of the PPF, I would like to pay tribute to them for the work that they have done in setting forth their ideas about risk and how to manage it. These ideas are particularly important to the Board of the PPF, as well as to the pensioners in this country, because they make our Board feel more comfortable about what may happen in the future and how we will deal with it. The PPF has a very important mission. There are pensioners all over the country who have worked hard throughout their whole lives with the hope that they will have a comfortable retirement with a good pension. In the days before the PPF it was not always thus.
We all know the PPF is now working well and is efficiently and effectively run. I am proud of our team. The best and the brightest of people are working at the PPF. It is real pleasure and an honour to be their chairman.
The Chairman: Thank you very much. I should now like to invite Mr Martin Clarke to introduce the paper.
Mr M. G. Clarke, F.I.A. introduced the paper. The full text of his introduction is given at the Edinburgh discussion on 20 January 2012.
Mr M. H. D. Kemp, F.I.A. (opening the discussion): Good evening, ladies and gentlemen. It is my pleasure to open the discussion of this evening's paper on the financial management of the UK PPF. As someone who actively encouraged the authors to commence this project, I am pleased to see it come to fruition quickly. I consider the outcome to be an impressive piece of work. My hope is that the paper will encourage others, including actuaries, to apply ERM in the wider pensions arena and not just to the special case of the PPF.
The UK Actuarial Profession has kicked off a strategic initiative recently to promote the expertise of actuaries in risk management and, in this context, it wants to build on its strengths in traditional areas such as pensions. It is great to have a paper like this that demonstrates in a tangible and coherent way the value that actuaries can bring to this area.
The aim of the opener on an occasion like this is to try to come up with a framework for debate. To do so, I should like to pose four questions, some perhaps addressable by the authors, but most aiming for a response from the floor.
The first question that I would like to pose is the extent to which the application of ERM disciplines to the PPF, as set out in this paper, form a helpful role model for the pension community more generally. The authors quoted, in section 1.1, from a paper that Chinu Patel and I presented to this chamber about a year ago that explored some of the challenges that were faced by pension schemes when applying ERM to themselves. One of those challenges, it seemed to us, was the relative lack of resources that particularly smaller pension schemes could apply to such disciplines.
It is clear from this paper that the PPF is unusual in the extent of resources it has available to apply to effective risk management. For example, in section 5.1.7 it is explained that the PPF audit and investment committees are supported by two executive committees, a risk management committee and an asset liability committee and also by the presence in the management team of a chief risk officer. The table in section 5.2.1 indicates that not only does it have an in-house investment team, but also an in-house risk team. So, maybe some of the disciplines that the PPF has been able to introduce are not implementable in practice by many schemes, particularly smaller ones. Then the question is: which are the most important risks for those schemes to focus on?
My second question is whether the floor thinks that the mix of ERM resources that is being applied to different sorts of risk by the PPF is optimal. Even though the PPF appears from this paper to have well-resourced risk activities, the resources it can apply to this task are, of course, not limitless.
Are there any gaps that you can see in what they are doing or any areas where you think that the effort being applied is disproportionate to the practical impact of the risk in question? Two possible areas that I noticed in this context related to the coverage of operational risk, which Mr Clarke mentioned earlier, and on which the PPF intends to focus to a greater extent as it becomes larger; and to reverse stress testing, which I will come back to in a moment.
My third question is: to what extent does the floor think that the PPF's risk and governance disciplines will prove robust in the longer term? The main reason for posing this question is, as Mr Clarke has already highlighted, that the European Insurance and Occupational Pensions Authority (EIOPA) has recently triggered considerable debate on how European Union pension schemes might be regulated in the future.
One of its proposals relates to the possibility of requiring pension schemes to construct so-called holistic balance sheets. These types of balance sheets aim to take into account the additional security provided to members from mechanisms such as sponsor covenants, conditional benefit structures, and, of particular relevance to us tonight, pension protection arrangements.
I am aware that EIOPA's proposals have generated heated debate, particularly here in the UK. Personally, I agree with the EU Commission in thinking that some of these reactions are a little excessive. It seems to me that the holistic balance sheet proposal provides a mechanism that allows pension schemes to have smaller amounts of tangible assets on their balance sheet than might otherwise be expected as long as they have other items that are off their accounting balance sheet, such as access to legally enforceable sponsor covenants or pension protection arrangements, that bring the level of benefit security they are providing back up to what is needed.
As UK pension schemes are relatively under-funded in many cases, but have access to these security proposals, you might have expected them to have welcomed this particular part of EIOPA's proposals. However, EIOPA is asking the industry for some objective justification that the status quo is acceptable. Who wants to justify what they are currently doing, especially if it is going to involve a lot of work?
The applicable pension protection arrangement in the UK is the PPF. The paper also summarises some similar pension protection arrangements operating elsewhere in the world which I hope will be useful for the 40% or so of the Institute and Faculty's membership that is based outside the UK.
Therefore, implicit within this holistic balance sheet concept is the need to form a view as to the reliability of the security underpins being offered by the PPF and its compatriots elsewhere in Europe.
To do this it is instructive to think laterally and to take an ultra-high-level overview of the entire pensions industry of a country. It does not necessarily have to be the UK. However, let us suppose that the country in question has a pension structure similar to the ones prevailing here. Suppose that at some stage in the future it runs into a situation where its pension schemes are in aggregate materially underfunded and, at the same time, its aggregate sponsor base supporting those schemes is too weak to afford to make much impact on the aggregate level of scheme underfunding.
What happens in this hypothetical scenario? What is the impact of including a pension protection arrangement such as the PPF?
In early years the pension protection arrangement would provide coverage, just as the PPF does at present. Beneficiaries of schemes whose sponsors defaulted in those years would be well covered. However, the schemes whose sponsors had not defaulted will be making levy payments to fund this support.
So, all other things being equal, their funding shortfalls would be increasing over time, as would the levies that they would be paying to the pension protection arrangement. Ultimately, the situation would become unsustainable, with schemes whose sponsors had not defaulted exhausting their asset base, in the extreme case, to support schemes whose sponsors had already defaulted. The lucky ones would then be the members of the schemes whose sponsors defaulted early on in the process. In practice, of course, something would give in the meantime. If this was ever the situation in the UK, then this might involve the PPF breaching the limit of its levy-raising powers or asking the Secretary of State to cut back the benefits it was covering.
But the key point I want to highlight is that inclusion of a pension protection arrangement might redistribute resources between schemes, but it would not solve a systemic problem if the problem was sufficiently large.
In this context, I would note that one aspect of modern ERM disciplines that did not seem to be covered in this paper was the concept of reverse stress testing. What this involves is adopting the hypothesis, however improbable, that the operating model of the organisation in question has stopped working and then brainstorming how this might have happened, and what risk mitigating strategies might have been implemented to make the outcome less unpalatable. Perhaps this is something that is already done by the PPF, and is just not shared with us in this paper, or perhaps we can do some of this brainstorming tonight if we think it appropriate.
My final question is: to what extent do we think pension schemes ought to be allowed to take into account the existence of pension protection mechanisms, such as the PPF, when assessing the level of benefit security being provided to their members. That sounds similar to my third question. The important difference is the inclusion of the word “ought” in the question.
Life is not as simple as we might expect. The UK has a conceptually similar protection arrangement applicable to insurance companies and banks called the Financial Services Compensation Scheme (FSCS). Regulatory solvency assessment of insurers and banks disregards the potential recourse that individual insurance policyholders or bank depositors have to the FSCS, or at least, it does so explicitly. Perhaps the recourse is taken into account implicitly in some way. This situation may be due to a conscious desire by regulators and politicians to limit what I would call “moral hazard”. By this I mean the possibility that shareholders of insurers or banks may otherwise be incentivised to pursue overly risky strategies because of the presence of lifeboat mechanisms such as the FSCS. Is this type of issue relevant for pensions? And, if so, what is the right balance to strike?
I would encourage us to explore the following topics. One, to what extent is ERM discipline being applied at the PPF a good role model for other organisations in the pensions community? Two, is the balance of resources being applied to different types of risk faced by the PPF appropriate? Three, how reliable are the underlying frameworks within which the PPF and its compatriots elsewhere operate? Four, to what extent is it appropriate, when assessing the benefit security being provided by a pension scheme to its members, to take into account the recourse the scheme or members might have to a pension protection arrangement such as the PPF?
However, I am conscious that there are plenty of other topics that we could also pursue. I would encourage you to contribute to the debate. It is in the wider interests of society that the best possible risk management disciplines be applied at the PPF and in the pension funds it covers. It is also in the particular interests of the significant proportion of the individuals here today who are members of the 6,550 or so pension schemes covered by the PPF, or who advise them, or who merely do business with them, or with the PPF itself.
Mr L. Churchill (guest, NEST chairman): Good evening, everybody, and thank you, Mrs Morgan, for inviting me to lead off in terms of the submissions from the floor. I have taken quite a close interest in the PPF over the years.
I will try to refer, Mr Kemp, to some of your provocations and questions, but you will forgive me if I do not discuss the analogy with the FSCS this evening.
There is a danger that telling the story to date makes the path that the PPF has been on seem easy. I can assure you that it has not been. The position today is the product of many years of highly skilled work in an environment of great uncertainty, a volatile industrial landscape and financial crisis. There were many challenges in shaping the final product, analogous, perhaps, to a sculptor creating a beautiful figure out of what started off as an amorphous mass of stone.
As the paper points out, the game is not yet over. The PPF remains a developing organisation. That said, if we regard the current position as similar to the first phase of the life cycle, where the priority was to avoid infant mortality and to establish the basis for a productive life, then the PPF is a success of which we should all be proud. I say “all” because of the contribution that many of you in this room have made to the PPF over the years.
The PPF did have highly skilled staff, and that was a necessary but not a sufficient condition of success. I would like to pick out a couple of contributions that people have made from hundreds of possible examples. I would pick out Andy Young's excellent and streetwise foundation in the Pensions Act 2004. I would pick out the role of external scrutinisers in the early days of the model development, checking that the model was fit for the purpose for which it was then being used. I would also mention the Actuarial Profession's own response to the initial consultation on incorporating investment strategy into the levy formula, pointing out some of the key faults in the suggested approach at that time.
The PPF's mind-set was never, “We have a problem to solve”. It was always, “There is a problem to solve and there are many intelligent people in the pensions community who can help solve it.” This inclusive approach and collective contribution gave PPF staff the best possible chance of success.
I applaud the adoption of best practice from adjacent fields that the authors have set out so cogently in the paper. The PPF may be unique, but it has several characteristics for which there is a good analogue elsewhere. So, the authors adopt best practice from the pensions field as if the PPF were a pensions scheme. They adopt best practice from the insurance field both in the creation of internal models and in the tests applied to calibrate the degree of security offered. They were early adopters of BAS's actuarial standards. Technical Actuarial Standards M: Modelling (TAS M) is specifically mentioned in the paper. The precepts of Technical Actuarial Standards R: Reporting Actuarial Information (TAS R), the reporting standard, were also vital in addressing the communication issues between the experts working on the details of the model, or on the investment and hedging strategy, and the governing body itself.
I should like to think that, in the true spirit of partnership, the PPF gave back for all that it received. For example, it created the concept of the Purple Book to enable data to replace anecdote as a basis for the evaluation of policy (a move, I must say, that was not universally welcomed by the Actuarial Profession). At a more technical level, it gave input to Chris Daykin and Chinu Patel's investigation of discount rates.
In addition, as a by-product of emerging insights from the risk-based model, the PPF was able to demonstrate that hundreds of millions of pounds of public money could be saved by changing the operation of the Financial Assistance Scheme.
The paper mentions that the time horizon to self-sufficiency is not set in concrete. I interpret some of Mr Kemp's remarks as saying the sooner the PPF is self-sufficient, the more resilient it will be in the face of some of the contingencies. This brings to mind the Aon Global Risk Survey for last year, which stated that 66% of schemes are targeting either self-sufficiency or buy-out, albeit without a strict definition of self-sufficiency.
In that regard, I think imitation is the sincerest form of flattery. People have been watching what the PPF has been doing with its liability-driven investment strategy and are minded to follow.
If this is borne out in practice, there is going to be a lot of risk mitigated over the next 10 years, giving the opportunity for the PPF to accelerate its own journey to self-sufficiency and hence to play a more significant role in the holistic balance sheets of UK pension schemes, as suggested by EIOPA.
I think, in theory, the concept of holistic balance sheets is fine. Mr Clarke has already mentioned the impact of quantitative easing tonight. This has the effect of increasing the off-balance sheet employer covenant at a cost of potentially reducing the on-balance sheet deficit. But of course one of the issues we are going to have with holistic balance sheets is how these elements should be calibrated. There is also the issue of how they can be applied. Are they better applied at the level of the nation, or can they be applied at the level of the individual scheme? There are many unanswered questions at the moment.
I think the PPF has a great role to play in driving the debate forward. The PPF is a new component of the UK's financial infrastructure. It has proven its worth so far, not only for its original function but for the Financial Assistance Scheme. So what else might it be able to do to make a contribution to the pensions industry? It is, in my view, a key part of the UK's answer to the EIOPA consultation on the revision of the Institutions for Occupational Retirement Provision (IORP) directive. If the UK is to mount a credible defence to capital requirements on ordinary businesses in the UK, then the strength of the PPF is a credible way to do it, and I should like to see it making a contribution to that debate.
I would also like to see its expertise in risk management applied to defined contribution pension schemes, to provide the much-needed security to millions more pensions savers. There is no reason in principle why, if you happen to be in a defined benefit scheme, there should be a higher degree of security than there would be if you were saving via defined contributions. I think the PPF's expertise here could give it an actuarial leadership role.
When I was appointed as the first PPF chair, an actuary wrote to me saying, “I always knew you were brave; I did not realise you were foolhardy. You do know that this scheme is doomed to fail.” I think this paper convincingly demonstrates how even adverse risks can be managed by rigorous application of actuarial science, and how the public interest can thereby be served. This combination of the public interest, actuarial discipline and risk management techniques is what can give actuaries a good name.
Mr S. P. Rees, F.I.A.: The creation of the PPF has affected UK pension provision in a number of ways and it is important for these to be debated by our profession.
There has already been debate over recent years regarding the levies set by the PPF. Most of that has been open and public. The level of debate regarding the PPF's own internal finances has inevitably been lower. The PPF has published some documents regarding its approach, but these have not been for the purposes of consultation or discussion; they have generally been announcements. So opportunities to discuss this element of the PPF's influence are to be welcomed.
I confess that I am a little disappointed by some elements of the approach taken in the paper. There is a tendency in some places to set a question, and then develop the appropriate answer to that question. I would have liked there to be more debate or commentary in some places regarding whether the question was the right one in the first place.
I have repeatedly made one particular point about the PPF's existence. This is the point about how the PPF was created by the Department for Work and Pensions (DWP). We know that the DWP was very careful when designing the PPF and that it gave the PPF some important risk-controlling features. In particular, it created the power for the PPF Board to amend the level of increases to compensation. It also created the power for the Secretary of State to amend the level of compensation itself. These powers cannot, of course, be used lightly and I am sure that there is a preference among all parties not to use them. However, they do exist and they exist specifically to change the way that risk needs to be managed by the PPF.
The paper acknowledges these powers in the last sentence of paragraph 2.9: “To date the PPF has not articulated the circumstances in which these powers might be exercised and for the purpose of its financial management and this paper such scenarios are not explicitly modelled”. I found this to be the most disappointing sentence of the paper because it is so important to the subject of managing the PPF's risks. I do hope that it will be possible for there to be debates in future that include this important feature of the PPF's structure so that a comprehensive analysis of the relevant risks can be attempted.
I also look forward to future discussion on what might be the right amount of risk for the PPF to manage. The paper gives commendable attention to ways in which risk can be removed, minimised or reduced. The PPF does, however, have a number of unique features that put it in a position where the positive attributes of risk could be actively embraced. It is not clear, for example, why the PPF should constrain itself with short-term risk measurement and budgeting in the ways that might be typical in other institutions. I look forward to more analysis of the right amount of risk so that those discussions can occur.
Finally, there is another important area of the PPF's management that needs attention. In section 3.5 it is acknowledged that the most likely scenario is for the PPF to be in surplus at some point in the future. I suspect that most businesses have a clear plan regarding their actions if events unfold in line with their most likely prediction. However, we know very little about which generation of levy payers will see the benefit of a surplus in the PPF, and in what manner. The lack of clarity in this area strikes me as an important area of risk for the PPF that might need more attention.
I am pleased that this paper has enabled the start of a debate regarding how the PPF works, and I look forward to future opportunities when some of the more fundamental points can be debated.
Mr A. J. Chambers, F.I.A.: I should like to commend the authors on producing an interesting and thought-provoking paper illustrating how the PPF has taken a thorough approach to analysing and managing both on- and off-balance sheet risks. While the PPF faces a unique set of circumstances, I hope it will encourage actuaries advising corporations and their pension funds to consider a holistic approach to risk management of their assets and liabilities.
I have a question about the detail of some of your analysis of the interaction between pension schemes and their sponsors’ deficits in different economic scenarios. Table 4.4 suggests that rising interest rates would result in an increased deficit. Is that because corporations are more sensitive to interest rate rises than their pension funds? Perhaps it could also be that, in your model, rising interest rates are bad for risky assets, that is, a negative impact from low economic growth, which we know pension funds are exposed to.
Mr C. A. Speed, F.F.A.: There are a few points which struck me as I read through the paper. Firstly, one of the key risks of the PPF is insolvency of a sponsor of a very large scheme. That is recognised, but there seems to be insufficient analysis of this. In section 3.5 there is discussion about longevity, and how this cannot be hedged, and unexpected claims. However, we are also told that there is a 10% margin to cover this. The paper also says that if you do a Solvency II type longevity stress, it requires an 11% margin. So implicitly we are being told that there is a -1% reserve for unexpected claims. This is an area in which I think a lot more work needs to be done and is a serious risk. Events such as Deep Water Horizon and the recent financial crisis suggest that some of the very large schemes are not as secure as we once thought.
Another area relates to the asset allocation. In the benchmark asset allocation, there is a 10% weighting towards equities. That is quite interesting because we know that many DB schemes have quite large weightings towards equities. These are mis-matches. The aim of the PPF is to be in a position after 2030 where they do not need further levies. So this mis-match, which already exists in pension schemes, seems to be exacerbated by the equity weighting within the benchmark portfolio.
If we delve into the points of the modelling, we can see that they are currently targeting an 80% chance of success after 2030, and an equity risk premium of between 
$$$3\tfrac{1}{2}\% $$$
and 4%. If I put these together it starts to look like an old-style pensions asset liability study. It effectively assumes equities outperform in the long-term. I found that slightly disappointing because Mr Clarke did make the point, and refers to it in section 7, that there is a risk of being misled by models.
Another area that struck me was that the smoothing of asset returns should have been brought out. To me, it seems like you are putting greater value upon historic information than more up-to-date information. And also the gap between RPI and CPI needs consideration. In the base case this is given as 1.1%, which is at the upper end of ranges that I have heard commented upon.
Finally, there is a lot of talk about setting the benchmark and the probability of success. I think we probably need some further work on the scenarios where those success criteria are not met. When we have a shortfall, how bad is it? What is the impact? What are the triggers?
It can seem depressing, but good risk management starts by analysing the situations where things go wrong, just like the reverse stress test to which Mr Kemp referred. This is another area in which I would like to see further work.
The Chairman: In his opening remarks, Mr Kemp posed four questions. I wonder if anyone has any thoughts, comments or responses to his first question, which is around the extent to which the application of ERM disciplines to the PPF are a helpful role model for the pension community as a whole. I work at the FSA helping to implement the Solvency II directive, which places an emphasis on governance. So I am interested to know what my pensions colleagues think about that. Do you have any comments or observations?
The second question was whether or not the mix of ERM resources applied to different types of risk by the PPF is optimal.
The third question was about whether or not the PPF risk and governance framework will prove to be robust. In respect of this we have had some comments about reverse stress testing.
The final question was about the allowance for the PPF in valuing pension fund liabilities and assets.
There have also been some comments about the use of internal models. I am again particularly interested in what pensions actuaries think about the use of internal models in valuing pension scheme assets and liabilities.
I throw that open to you who are more expert than I am in pension funds.
Mr D. P. Benstead, F.I.A. (in a written contribution that was read to the meeting by Mrs Morgan): The PPF has the potential to be the making or the breaking of the defined benefits pension system in the UK. I was disappointed that an 85 page paper on the management of the PPF did not address itself to which of these it is.
The PPF levy is highly loaded against the schemes with lesser funding and/or weaker employers. The maximum levy cap of variously 0.5%, 0.75% and 1.0% of liabilities per annum equates to 10% to 15% to 20% of liabilities over the lifetime of a scheme. A 1/60 scheme, left to run for long enough, can build up liabilities of ten times the active members’ pensionable salaries. A levy of 0.5% to 1% of liabilities equates to 5% to 10% of salaries per annum, just on the levy. We cannot expect employers and schemes to survive under that weight of levy. Even if the average levy is sustainable, this loading of levy onto a minority of schemes has the capacity to break the system by levying schemes one by one into the PPF, weakest first.
It is important to remember that the levies collected from schemes which later join the PPF are of no value: the PPF gets all the assets anyway. It makes no difference whether the PPF collects 1 per annum over the previous 10 years and 90 when a scheme joins the PPF, or nothing per annum over the previous 10 years and 100 when the scheme joins the PPF. The only levies that matter are the ones collected from schemes that don't join the PPF. For a sustainable system, the levies should be spread as thinly and evenly as possible across all schemes. I know that the law says that levies must be 80% risk-based, but that does not mean the scales have to be as heavily tilted against the weakest schemes, or the ones with the minimum pension increase rules, as they are.
We must not assume that employers have bottomless pockets which can be endlessly tapped by retrospective benefit improvements, retrospective raising of the cost per £1 of pension on a scheme wind-up, loss of tax credits on income on existing assets, de-risking (equals expected cost increasing) measures or a bottomless pocket to pay the PPF levy. The PPF must not fall into the same trap as the politicians have down the years.
Imagine a system in which all schemes where the employer suffers an insolvency event join the PPF. If the PPF seeks to fund itself to a higher standard than the average funding of schemes outside the PPF, then the stage is set for a futile vicious circle. The more schemes that join the PPF, the more levies the PPF raises to improve its own funding to its target, the more employers cannot afford their schemes, and the more schemes join the PPF whether by pre-pack administration or other insolvency event at the employer. When the PPF has finished collecting all schemes into itself, it will not have achieved its funding target; it will, of course, have the same funding as was the average across all schemes before it collected them all into the PPF.
Of course, the PPF does not take all schemes, only the worst funded ones, so there is no cross-subsidy from the well-funded ones to the worst-funded ones. This makes it all the more important that the PPF's funding target is not too high. At most, the PPF should raise no more levies than required to raise its funding level to the average of schemes outside the PPF. Ideally it sets a lower target, to increase the sustainability of the system.
The product of PPF compensation and the expected cost per £1 of pension provision in the PPF is, to my mind, too high. PPF compensation in respect of those schemes which have only ever given statutory minimum pension increases in payment is only slightly less than the scheme's benefit. The product of PPF compensation and the cost per £1 pension of its current strategy is likely to be rather higher for these schemes than a reasonable ongoing funding target. These schemes are particularly vulnerable to an excessive levy.
I am disappointed that the paper, to the little extent that it does identify structural problems, does not proceed to look for sustainable answers.
For example, in paragraph 2.12, it is observed that the supply of schemes to levy might run out and the “solution” to this problem is to get in quick and charge levies while the PPF can. Such behaviour only helps to exaggerate the problem by making DB schemes less attractive or even unaffordable to sponsor. Potentially better solutions include lobbying the powers that be for a DB system in which employers are pleased to participate, or lobbying for lower PPF compensation, or seeking to provide the existing compensation at a lower expected cost by using a less cautious investment strategy. Surely the PPF has influence. Let's see the PPF trying to be a part of the solution, not a part of the problem.
It can be argued that the politicians knew that setting Section 75 debts at the buy-out deficit was an unaffordable funding standard for employers and created the PPF as their solution to the problem of pension provision in the event of a scheme wind-up (paragraph 2.3). If so, then the PPF behaving somewhat like an insurance company merely reinvents the problem it was created to solve.
Planning to pay a pension promise in, say, 60 years’ time “without the need to take future risk” (paragraph 3.3.1) is not necessarily an appropriate objective. Of course there is risk in a system whose task is to plan to make payments decades later. Wholly eliminating risk over that timescale is unlikely to be achievable. Attempting it will create a high cost system which is still vulnerable to failure due to events we did not foresee, or could foresee but could do nothing about. Paragraph 3.3.3 worries that there might not be schemes to levy in future; I worry that levying excessively now helps ensure that future comes about. It would be better to work towards an environment in which DB schemes have a long term future.
The paper notes, in paragraph 2.9, that varying PPF compensation might be possible but does not explore the idea. Rather than rigidly fixing PPF compensation decades in advance, and therefore rooting compensation in a standard of living which may be decades out of date by the time it comes into payment, and achieving that small thing at high cost due to the low return investments required to match those obligations, it could be better to design PPF compensation more flexibly and take advantage of that to adopt a more rewarding investment strategy.
To construct a very simple example, if the PPF invested exclusively in UK equities and related annual increases in its compensation to UK equity dividend growth we would have a system which, in 60 years’ time, paid compensation which is proportionate to the standard of living in 60 years’ time, and is affordable in 60 years’ time. Designing the liabilities to match “growth” assets rather than designing bond assets to match fixed liabilities has the potential to be a better system and is worth exploring.
So, to come back to my opening question, will the PPF be the making or the breaking the defined benefits pension system in the UK? I perceive structural problems that mean the PPF could, if the course is not altered, be the breaking of the system. The levy cap is too high, and the levy should be more evenly spread both across schemes and across time. The product of PPF compensation and the cost per £1 of its provision is too high, especially for schemes which only give statutory minimum pension increases.
It is within the PPF's power to levy more appropriately and to work within the limitations of a sustainable funding target. It should do both. Taking the current definition of PPF compensation as a given, that is likely to mean endeavouring to provide compensation at a lower expected cost per £1.
There is a debate to be had on whether the PPF should, within a non-burdensome funding target, seek to provide a small, tightly defined benefit from a low investment risk and high cost investment strategy, or whether it should provide more flexible benefits as can be afforded from a “growth” asset investment strategy. Or whether it should rely on the long term existence of DB schemes because, surely, we should all be working to achieve that. I don't mind which, so long as the outcome is a PPF that is not a system breaking burden.
Mr W. Rayner, F.I.A.: The paper is a helpful record of how the PPF operates and of the risk management techniques used. In particular, I applaud the sentiments in section 10.2. This commentary emphasises the need for an unambiguous linkage between business strategy and a technical analysis, clear delegation of decision-making and accountability and an optimal combination of good communication and understanding on the part of the decision-makers as vital for making appropriately well-informed judgements.
The authors describe governance as a work in progress, and while, no doubt, that is a statement of fact, I suspect they might be being unduly modest. Something new, whether it be a law, a problem, the unexpected, or a new personality on the board, inevitably comes along at some point to challenge the status quo. Sometimes these challenges are valid, sometimes not. Achieving good governance is, in my view, always a work in progress.
It is the specifics of securing benefits on which I would like to focus. I state no personal preference of covenant versus capital as a means of securing liabilities. In section 8.2, the paper does take a sideways look at how the PPF levy would be priced according to insurance pricing principles. Also, reference is made in paragraph 3.5.10 to the PPF's target funding margin of 10%, comparing that with the impact on PPF liabilities of one of the stress tests applied to insurance companies under Solvency II. As I think we are all realising from the discussion tonight, Solvency II for pensions is a topic fast approaching, if indeed not already here.
If, for the sake of illustration, I were the sole proprietor of a very large enterprise which sponsors a defined benefit pension scheme I would not be enthusiastic about the prospect, as is suggested in some quarters, of setting aside capital as a buffer to secure members’ pensions. I would want to put that capital to much better use in my business. However, I also want my employees’ pensions to be secure. That is quite a dilemma.
Asking every sponsor of a defined benefit scheme to hold capital as a buffer seems to me like a very large buffer indeed, arguably too big a buffer in the context of defined benefit schemes nationally. I consider my illustrative business large enough to arrange pensions for my employees without financing via any insurance policy, and consequently via the insurance company's capital buffer.
What about an alternative of holding aggregated capital on behalf of all defined benefit schemes as an aggregate buffer? Instinctively, and recognising that I have not done any analysis, I would expect the aggregate capital to be less than the sum of the individual scheme buffers to provide the same level of protection for members, possibly significantly less. I would like the authors to comment in their summing up on whether the PPF's target funding margin of 10% might be a proxy for capital held to secure the safety net of the nation's defined pension benefits, albeit only up to the PPF compensation limits.
I note the reference in the appendix to international comparisons and suspect that the issue of capital buffers will also be relevant to counterpart organisations based in Europe.
Mr R. S. Bowie, F.F.A.: First, thank you to the authors for the paper. Lady Judge mentioned the warm relationship of the PPF with the Institute and Faculty, which is reciprocated.
Our initial work with the PPF had some creative tension in it as we worked through the operational interface between us, but latterly that relationship has been maturing. I think the fact that you are prepared to bring this paper to one of these meetings is part of that maturing relationship. It is enormously helpful in informing the Institute and Faculty's thinking on public affairs.
That has had an important recent impact in our response to the EIOPA consultation, of which something was said earlier. If you have not read the Institute and Faculty response I commend it to you. We supported the concept of a holistic balance sheet and could find no sensible counter argument to that. It seemed self-evident that if you were giving financial undertakings to individuals, it makes complete sense to put it in the context of a holistic balance sheet. But we were careful not to allow that to be read directly across into a pensions equivalent of Solvency II for a number of reasons. I will mention two here.
First, the legislative promises that have been made in the pension scheme are fundamentally different from those made in an insurance contract. That is true across Europe with the possible exception of France. Second, it seemed crazy to us to introduce anything that looked like Solvency II to pension schemes when a much better resourced and governed insurance industry is still struggling with the practical consequences.
While we welcome the holistic balance sheet proposed in the consultation, we cautioned against the detail, and strongly urged EIOPA to recommend full impact assessment and cost benefit analysis.
If you look on the EIOPA website, you will see that they have specifically responded to the Institute and Faculty's comments, and we are pleased to see that they were received constructively.
The clear steer from EIOPA that there should be an impact assessment is one that we welcomed. Not all of our recommendations and not all of the responses were so enthusiastically taken on board. But, nonetheless, we believe that we made a mature and reasoned contribution which is not something that every UK respondent seems able to have done.
I seek to answer three of your four questions, Mr Kemp, if I may. Firstly, I believe that the PPF is a model of how that holistic balance sheet could and should work in practice. You asked about anything that may have been missed out. Whereas, they have not been missed out, I think the role of governance has been underplayed, as has the role of scenario testing. Both have been mentioned in the paper, but much less than modelling. There is always a risk that, when the bulk of the paper refers to the results of modelling, the careless reader might be tempted into the trap that all problems can be solved by modelling.
To my knowledge, the approach is being used by forward-thinking UK funds, and a number of people in the audience will have had meetings with the Pensions Regulator over the course of the last few weeks. The Pensions Regulator's approach to its stewardship of UK pension schemes is in tune with what you are doing.
That takes me to one of the points that is not mentioned. I know that the PPF is careful not to mention the Pensions Regulator, and vice versa. But one of the risk management tools that you are not allowed to mention is the fact that the Pensions Regulator is trying to avoid pension schemes being in a position where if a company failed it would require the PPF. It is slightly disingenuous in the UK framework for the benefit of working together not to be mentioned in the paper. Therefore my plea to the PPF would be to gain some clarity about that relationship.
Mr Kemp, you asked the question: is this robust, and how would we take into account the role of the PPF? If the UK does not get its act together and suggest a way in which the PPF is included in that holistic balance sheet, then EIOPA, or some other part of the European Commission may force one on us.
Mr H. D. Sutherland, F.I.A.: I would like to commend the paper. I hope it will be regarded not only as a useful contribution to the profession, but that students would see this as a useful case study.
With that in mind, can I just bring up one minor typo in table 4.3? In the third item, the scenario is a sharp rise in bond yields, but the description talks about economic uncertainties for major economies resulting in an unsustainable rise in bond prices.
A bigger issue is how this work, and the use of modelling, is communicated to the boards of the organisations concerned, and more generally. I fear that there is a danger that people will start to say, “The actuaries have done this work. They have crunched the numbers. Here is the result, and we are going to live with it.”
The paper refers in a number of places to the need for the limitations of the model to be communicated and to training board members in how to use and interpret stochastic models. I think this is great but I wonder whether the ability to do it is particular to the board structure of the PPF and whether it would be equally feasible for other financial institutions. I think that there is a danger that the model is simply adopted and maybe that would cause the profession problems in the future.
Mr J. M. Lowes, F.I.A. (in a written contribution that was read to the meeting by Mrs Morgan): The paper explains the PPF's intention to be fully funded on a self-sufficiency basis by 2030, having eliminated the on-balance sheet risk as far as practicable. This reflects the reduced capacity to eliminate any PPF deficit thereafter because of the anticipated reduction in the PPF universe and in the risk represented by that universe. Section 9 of the paper explains the investment strategy including the investment return target and associated risk budget. So my question for the authors is whether the intention is:
(A) to maintain the investment risk budget as a percentage of assets over the period to 2030, with a cliff edge reduction in 2030 (or earlier if the target funding level is reached earlier); or
(B) to maintain the investment risk budget in pound terms over the period to 2030, which implies a gradually reducing return target over that period (due to spreading a flat pound risk budget over an increasing asset value), with a cliff edge reduction in 2030 (or earlier if the target funding level is reached earlier); or
(C) to reduce the investment risk budget in pound terms over the period to 2030 towards the target of nil by 2030, which implies a return target which reduces more rapidly over that period (due to spreading a reducing pound risk budget over an increasing asset value).
The second and the third approaches imply a reducing target for the outperformance of assets over LIBOR over the period to 2030. How has this been taken into account in the modelling of the probability of successfully achieving the PPF's financial objective by 2030?
Mr G. P. Everness, F.I.A.(closing the discussion): I think there are various reasons why a paper like this is particularly to be encouraged. First, it is good to have a paper on a subject that crosses several different practices. From what I heard about the discussion in Edinburgh, there was a range of contributions from investment, insurance, pensions, and risk management.
Second, even though the authors point out right at the beginning of the paper that the PPF is a unique institution, their risk management ideas and processes are potentially applicable to other pension schemes. Mr Kemp wrote a paper a year ago on this subject. I am sure that this is an area which needs more consideration from pension schemes.
Third, the PPF is a substantial public body. A significant proportion of the population has a direct stake in it as do many large employers. So it is very much in the public interest that its financial management is properly scrutinised. We have heard testimony tonight from both the PPF and the profession about how it has been good to work together to try to reach the best solution. With the best will in the world, nobody has all the answers on these matters and so what is needed is a process with good scrutiny and challenge to address them. Where people have said that they do not agree with this or they do not agree with that, the PPF people, I am sure, will not take it as a criticism of the way in which they are going about their work, but rather as an endorsement of the importance of what they are doing and the need for this sort of debate to take place, in order to work out the best solution.
We had the four questions posed by Mr Kemp but, as he predicted, the discussion did not always refer to them. The first of them was about the PPF approach being applied by pension schemes generally. We did not hear much about that, but we did have a comment to the effect that for defined contribution as well as defined benefit schemes there was a need for this type of holistic risk management. Maybe that is an area for further investigation.
Then we had the question relating to whether the PPF had got the balance right as regards the attention given to different types of risk.
At this point, I will throw in a particular comment of my own which relates to what risks the PPF particularly wants to focus on and why. I think I understand why it is focusing on the particular type of risk on which it is. It comes largely from its mission statement, which is “to pay the right people the right amount at the right time”. This is focused on the first of the stakeholders I referred to – the members of pension schemes who are covered by the PPF. There is not any mention in the mission statement about levying the right schemes the right amount at the right time.
I can understand that, but I still want to explore the fact that there are some risks involved in relation to the correct levy collection as well as in relation to paying the right benefits. In particular, it seems almost inevitable from the idea of becoming self-sufficient by 2030 that there will be some significant element of cross-subsidy between different generations of scheme sponsor. We could debate at length about whether or not the levy achieves the right distribution between companies on a year-by-year basis. But it seems that, if there is a desire to de-risk as quickly as possible, there must be a substantial risk that there is a clear cross-subsidy with overcharging in the short term.
I am not saying that this is definitely the wrong way to operate; only that it is a point that should be debated.
Relevant to this, the question of surplus was raised by at least one person. It is true that any strategy which is designed to minimise risk in terms of paying out the right benefits is almost inevitably going to have some risk on the other side. Given this, there does then need to be a policy as to what is going to happen to any surplus. Otherwise, with schemes paying higher levies than might be expected on a best estimate basis, or other basis that is not as low-risk as that of the PPF, those schemes might be inclined to wonder why they are doing so and who is going benefit in the end.
Looking at the figures in the sensitivity table in paragraph 4.1.6, it appears that the first hundred million pound reduction in levy only makes a 2% difference to the probability of success. The amount of the levy, which is what matters to schemes paying it, does not seem to be making much difference to the probability of success. Although figures are not quoted, the table suggests, on an extrapolation basis, that it may well be the case that there is a greater than 50-50 chance of no more levies being needed. I am not sure whether that is true. This illustrates the variability in the possible levy outcome, and the need to understand what will happen if there is too much money left at the end.
Let's move on to question three, which I would summarise brutally as: “Is it going to work?” There were some quite different responses to this. Some people said that if we want to make sure that it works, we need to minimise the risk even quicker, and move earlier to self-sufficiency.
Then we had another speaker talking about the fact that the risk of insolvency for the largest schemes had not perhaps been taken properly into account, and that the PPF might be taking a bit too much equity risk. There were therefore arguments in favour of reducing risk and becoming self-sufficient quicker. On the other hand, we had other contributions suggesting that attempts to reduce risk too quickly would end up not only costing too much, but also potentially bringing the PPF down by putting too much burden on schemes.
Perhaps one of the points fundamental to where one might sit on this spectrum is the question that a couple of people have raised relating to the power to adjust levels of compensation, in particular, the revaluation and indexation. There are also powers to reduce the benefits more generally, but I think we can focus first on the fact that revaluation and indexation can be reduced, but can also be increased, that is, the power can be used either way. This leads to the suggestion that the PPF should perhaps be more willing to take a line which treats these increases as a variable part of the package, so that more risk can be taken for the greater good of potentially providing ultimately higher benefits, albeit that these benefits are less certain.
Also on the “Is it all going to work?” point, we had various comments about models. There is a general concern that anything based on a model could go wrong, which of course is possible. The long-term risk model appears to be central to the PPF's strategy. Clearly, one of the risks is that the model does not work, is inappropriate, or goes wrong. I was not quite sure to what extent the risk that the model is itself wrong was part of the risk management structure that the PPF has put together.
Mr Kemp's final question related to how the existence of the PPF might be taken into account in financial and governance strategies.
The thought that first occurred to me was that the Ilford case a couple of years ago established that the existence of the PPF should not be a relevant factor where trustees seek to take advantage of it as a justification for acting in a way that would otherwise be improper. However, without wishing to pretend to be a legal expert, that case was extreme, and it seemed fairly obvious that it should be considered improper.
When we are talking about the holistic balance sheet, including the Solvency II proposals that were mentioned, we are essentially considering to what extent the existence of the PPF should be allowed for in a funding basis. A speaker towards the end said that if our line is that, if and when Solvency II comes in with its holistic balance sheet, the PPF should be taken into account, we need to decide exactly how this should be done in the UK pensions context.
The first thing that occurs to me is that the PPF is not covering all the benefits. There is a slice of benefits at the top that is not covered by the PPF. This slice is also at the bottom of the priority order, where it is obviously less supported by the employer covenant. How is this slice covered in a holistic balance sheet? The point is that we need to think carefully about how the PPF would be brought into such a balance sheet.
Mr S. R. Rice, F.I.A. (responding to the discussion on behalf of the authors): Mr Kemp, in his opening address, spoke about the assumption that UK pension scheme funding will, in the long term, improve on account of the efforts of trustees, sponsors and the Pensions Regulator. “Will our funding strategy fail”, he asked, “if this assumption does not come good?”
What I would say to that is our funding strategy is indeed dependent on the assumption that our risk, that is, deficits in pension schemes in existence, will have largely disappeared by 2030.
Our belief is that UK pension scheme funding will, in the long term, improve on account of the efforts of trustees, sponsors and the Pensions Regulator. But if, as we journey towards 2030, that assumption looks unlikely to come to pass, then that is not especially difficult for us to manage. We can just shift our funding horizon. So if, ahead of 2030, we think that by the year 2030 not all the risk will have gone, that is, we can see that there are still going to be some significant deficits in that year, then we can simply push out the funding horizon to 2035, say. Or we might do the reverse. If, as time goes by, we thought that the risk was going to substantially disappear by 2025 then we could make that our funding horizon instead. We agree that 2030 is an important assumption and we are expecting the Regulator to be working with sponsors and trustees to de-risk. We also believe at the moment that de-risking will largely have taken place by 2030, but we can be flexible.
There were one or two questions about short-term risk and whether there was any particular need for the PPF to operate with significant regard to short-term risk? My response to this is that, whilst it might theoretically be sound to just focus on the long term, in regard to the PPF, in practice there is a problem. Members and levy payers have, for better or worse, a short-term perspective, and we, the PPF, would not wish them to have a bumpy and fear-inducing ride on our journey towards our long-term funding target of self-sufficiency in the year 2030. We try to show consideration for how our stakeholders might feel. I think that there is also a smaller and secondary reason: that it is not inconceivable that there might in the future be a big change in the UK pension protection regime. To be clear, I have no reason to suppose that this will happen, but if there were to be a change, then it is conceivable that the PPF might be put into run-off, and suddenly the short-term funding position does become very important. There is an analogy with pension schemes.
Another question that cropped up was about what the PPF was going to do with any surplus that existed beyond our funding horizon in 2030. What would we do if we ever reached the point where we almost certainly had more money than we needed to pay our liabilities? This is a difficult question and I do not see any particular guidance in the legislation regarding it.
If there is a surplus existing beyond the year 2030, then I think it is likely to have arisen because some of our 10% funding buffer, which is part of our self-sufficiency funding target, will have proved to be unnecessary. The buffer is largely there to cover longevity increasing at a faster rate beyond 2030 than we are expecting.
So what can we do to manage the risk of stranded surplus? One of the things that we are doing is to look at whether we might be able to hedge or insure our longevity risk in the future. If we can, and we can do it sizeably, then we will not need a buffer as large as 10% and the risk of a big stranded surplus is diminished. We would not remove the problem altogether, but at least we would be managing down the size of the potential stranded surplus. So we are addressing the ultimate surplus issue, just as we are addressing the issue of an ultimate deficit.
There were a number of questions and comments about when the PPF Board might think about cutting compensation. The comment was made that the Board has the power to cut indexation and revaluation of PPF compensation. The Board also has the ability to recommend to the Secretary of State adjustment of the 90% and 100% levels of PPF compensation, one or other of which applies according to whether the scheme sponsor's insolvency date is before or after the member's normal pension age. Now it is true to say that the Board has not articulated the particular circumstances in which it would seek to adjust compensation. As a result of that, we have not made much progress with the reverse stress tests that Mr Kemp mentioned. But what I would say about adjusting compensation is that it is likely to arise in extreme circumstances such as those occasions when our funding strategy has failed. The backdrop to this failure would probably be some nasty economic circumstances and probably difficulties for employers in being able to sustain the funding of their pension schemes. The Board's response to these extreme circumstances would be very difficult to model. For example, think back to the 2008-9 economic crisis that engulfed this country. It would have been difficult to have modelled that crisis before 2008. What might have happened with such modelling? We might have foreseen that something bad was coming, but who, for example, would have predicted that Northern Rock, Lloyds TSB and RBS would have encountered the difficulties that they did encounter and, secondly, that the Government would step in and engineer rescues for those companies, meaning that their pension schemes would not suffer insolvency events. It's hard to predict the outcome of future difficult economic conditions, and therefore it is very difficult for the Board to say in advance what action, if any, that it would take in those difficult conditions. So, of course, it follows that it would be difficult to model appropriate compensation adjustments in difficult economic circumstances. I don't think it would be very helpful or illuminating either.
What we can do is to concentrate on improving our probability of success, the target for which is 80% at the moment. We should focus on the scenarios that presently lead to funding strategy failure and try to turn them into funding strategy success by means of clever investment strategy, hedging, insurance, derivatives etc.
There have been a few questions about our attitude to some sort of Solvency II regime for pension schemes. Such a regime might seem to make things rather better for the PPF because there should be better funded schemes. Even so, we are cautious. There is a risk that, if it were implemented too suddenly, there might be schemes whose sponsors could have recovered from a low point over a period of years suddenly being catapulted into the PPF. Also sponsors may cease to provide DB accrual for their members. Neither of these eventualities would be good for UK pensions as a whole. All in all, we are somewhat ambivalent about Solvency II for pension schemes. We are, nonetheless, contributing to the debate. Our Chief Executive will be contributing to one of the European Commission's debating forums to share the PPF's perspective.
In conclusion, I should like to thank everybody who has contributed so thoughtfully and valuably to the discussion. We, the authors, have found that adopting a risk framework and quantifying risk have helped the PPF in the management of its risk. Our hope is that the paper and the discussion will prove to be a positive contribution to the theory and practice of enterprise risk management.
The Chair: Thank you, Mr Rice. It remains for me to express my own thanks and I am sure the thanks of all of us, to the authors, the opener and closer and everyone who has participated in the evening's discussion.
