The comparative value of pensions

OF THE DISCUSSION The author (presenting the paper): This evening we have before us the Scott Report and, in expansion of the technical side, my memorandum, which is basically the evidence that my Department gave to the Scott Committee when it was sitting, so it covers the points that were at issue 15 months ago. This subject leads to the question: if a comparison is made between the total package of remuneration of two jobs for guidance on the pay scales of one of them, how should allowance be made for pensions? Controversy centred around the assessment of the value of Civil Service pensions, which is a job I did on a joint remit from the Government as employer and from the Civil Service unions. The remit being directed to myself as an actuary, I always felt it essential to stick to actuarial methods as commonly used in the profession. To do anything else would not seem consistent with my responsibility towards whichever party was disadvantaged by the change. The most important event since Scott reported is the issue of indexed gilts, which is not dealt with in the papers before you. While it is too early yet to say how they will settle down, the market experience so far, which is of real yields between 2 and 3%, just cannot be brushed aside. We do not know how a generally available issue would go, but there is no doubt that pension investment is the right field to test for our purposes, and pension funds, with their final salary liabilities, require an indexed bond. There is no mention of private sector practice in this field, but there are management consultants who advise on pay scales. They make job comparisons of various sorts and some, I believe, build up data banks. I know some employ actuaries. Pensions must come into these comparisons, but I cannot find anything published about private sector practice. It would be valuable to hear something about how private sector methods compare with those we use in the public service. Mr R. G. Brown (opening the discussion): We are concerned with the valuation of index-linked pensions against the background of the salary comparability exercise. This subject has been considerably clouded by a large number of ill-informed comments or just plain misunderstandings. The main misunderstanding has been that the 'Deduction' recommended by the Government Actuary represents the cost of the index-linking. This misunderstanding is clearly demonstrated by the evidence of the Committee for Policy Studies. The purpose of the exercise is to evaluate the difference between the benefits of the public sector scheme concerned and the analogues. How this should be done is the central issue for tonight's discussion. I think the three key questions for consideration are: (1) the choice of the analogues; (2) consideration of the question of uncertainty; and (3) the assumptions used in the valuation. Since the publication of the Scott Report, we have seen the first two issues of index-linked gilts and these do help a little when considering some of these key questions. Dealing first with what is probably the least controversial of the three points the choice of the analogues I had not realized until the publication of the Scott Report that the analogues include public sector schemes. The pensions' calculations are part of a wider salary comparability study and, clearly, if public sector schemes are included in the analogues for basic salary comparison, then they must also be included for pension purposes. However, the inclusion of public sector analogues does introduce circular arguments and must inevitably lead to a degree of public disquiet. I believe Scott's recommendations to exclude public sector analogues from the salary comparability study to be