Imperial College London challenges the USS methodology

Members might be interested to read Imperial College London’s response to UUK on USS’s consultation on the proposed assumptions for the scheme’s technical provisions and recovery plan


We are writing in response to your request of 4 November to provide feedback on the USS consultation on the assumptions used for the technical provisions and recovery plan and in particular to comment on your proposed response that was circulated on 21 November. We are publishing this response on our website and will be sharing it directly with USS.
We are disappointed and concerned by your proposed response.
We are disappointed that you appear to be focused on trying to fit your current proposed benefit solution to the perceived problem without first sufficiently challenging all the assumptions.
We are concerned that without this challenge you risk recommending a major downgrading of one of our employees’ most important benefits based on numbers which are as likely to be modelling artefacts as a reflection of the underlying economic reality.
You raise several good points about the assumptions used to set various modelling parameters; we provide more supporting detail below, however we believe that this is in many ways a second-order issue and you are missing the more important point of whether the output as presented can be relied upon.
Some of our academics with significant expertise in relevant fields have been analysing the assumptions and modelling approach used. The model they have used is, by their own admission, highly simplified; however when calibrated against the cases put forward in the USS consultation document it nevertheless gives a good approximation to the funding levels. This leads us to believe that the key conclusions that can be drawn from it are unlikely to be contradicted by the full model that the scheme’s actuaries will have used. As you will see in the attached paper, our simplified model includes additional cases to those presented in the USS paper, and those that have used the actual salary increments and investment performance of the last decade show the model fund in surplus.
This model demonstrates that the USS data are just showing how sensitive the actuarial model is to the input assumptions, rather than illustrating the volatility of the actual deficit itself. We note that so far USS have felt unable to respond to our request for more transparency.
We therefore strongly believe that at this point USS have not provided sufficient information for UUK to make any recommendations and that UUK must insist that USS step back and provide full transparency on its modelling approach and the sensitivity to the changes in assumptions. This point is crucial because only once we all understand the genuine risks we face will we be in a position to move on to the debate on changes to benefits.
Linked to this, now the potential impact of this reduction in volatility on the long-term cost and benefits available within the scheme is starting to become clearer, we believe that the distinction between achieving an average of 18% employer contribution over the long-
term versus the variability around this average in the short-term needs to be revisited as matter of urgency.
We acknowledge that many institutions have expressed a desire to reduce volatility within the scheme; our institution is not one that shares that view. We recognise that this is outside the scope of the consultation on technical provisions, but believe that it is essential that UUK in its role as the representative of all its members starts further discussions on this point in the very near future.
Moving onto the more specific issues, we fully support the point you make about the draft assumptions made by USS being excessively prudent. Frankly, we feel this point should be made more forcefully in the final response. The assumption of RPI+1% for a general salary increase, with additional increases on top on an age-related scale, is far removed from the experience within the sector in recent times and not something, unfortunately, that we feel can be reasonably envisaged going forwards given the financial constraints within which we need to operate. We have looked back at our own data within College over the period since 2001 and this would indicate that an assumption of RPI was more reasonable than RPI + 1%, without any additional, age-related, increments on top. It should also be noted that our College has not been part of the national wage bargaining process since 2004 and our salary increases have never been lower than the national agreement and have generally been above, reflecting the higher cost of living in London.
We feel you make an important point in your note of 21 November about how the discount rate proposed by the trustee does not reflect any deterioration in either market conditions or the strength of the employer’s covenant compared to three years ago, but rather a change in approach to risk by the trustees, which might be influenced at least in part by their perception of the Pension Regulator’s position.
We agree with the view that gilt yields are artificially low at the moment as a result of the Quantitative Easing programme and feel much more economic analysis is required to support the assumption put forward by USS. We believe the requirement for an appropriate level of consistency between the assumptions related to the assets and liabilities needs to be stressed, as in the current proposal the former is too prudent and the latter overly pessimistic.
We also support UUK’s position that the recovery period should be 20 rather than 15 years. The arguments you make about this are sound, and we believe USS are being unreasonable in seeking the extra prudence, on top of all the other pessimistic assumptions, of keeping to 15 years
We hope this note makes clear how important it is for UUK to obtain more details of the full range of scenarios that have been explored and the sensitivity analysis that has been undertaken. We appreciate the challenge of ascribing appropriate probability distributions to the various variables and determining the correlations between them; having knowledge of these would help us give a more informed view on the appropriateness of the choices made. USS notes in its consultation paper that: “Continuing engagement with the scheme’s stakeholders is important as the valuation progresses and the trustee envisages further consultation and engagement with both employers and members, and their representatives. The trustee wishes to be as transparent as possible about its approach to scheme funding.”
We have tried, unsuccessfully, to obtain the additional level of detail we feel is required on confidences intervals, error bars etc. to give a sense of the range of estimates; we feel the scale of the changes being contemplated fully justifies this being made available for consideration and comment.
In summary, by saying that you do not expect the trustees to respond to all of the points made on specific elements of the technical provisions, but rather to look at the overall level of prudence, gives the impression that you are limiting the scope of your response to establishing that UUK’s current proposed benefit reforms are affordable. This is inappropriate given the points we have made about the sensitivity of the model to assumptions. We feel it is important to focus on justifying appropriate distributions for each assumption before a fuller range of scenarios is tested. We encourage a more open debate at this level of detail; otherwise our members will feel that change is being imposed upon them because of the modelling approach and the way the Pensions Regulator considers schemes in general, rather than specific factors related to the USS scheme. This is likely to lead to more substantial industrial relations impact across the sector in the New Year.
Most importantly, we risk not fulfilling our duty as employers to think through clearly and carefully, with all the data in front of us, one of the most important decisions we will have to make about how we treat our academic community. We strongly urge UUK to ensure USS makes available the information needed to make the right decisions

Leave a Reply

Your email address will not be published. Required fields are marked *