[Skip to content]

Sign up for our daily newsletter
The Actuary The magazine of the Institute & Faculty of Actuaries
.

Pensions Board news

What should MFR look like?

There has been a lot of discussion recently among pensions actuaries about the future of the minimum funding requirement (MFR). The purpose of this article is not to add to that debate but to consider what the original objectives of MFR were and what MFR would look like if it were being set to meet these objectives in current economic conditions. What were the original objectives?

  • With the exception of large schemes, pensioner liabilities were to be valued broadly in line with the approximate cost of buying out the liabilities with an insurance company. For this purpose, it was assumed to be the approximate cost of buying out members that were blue-collar workers as opposed to white-collar workers.
  • The value of the non-pensioner liability was intended to give a fair chance of being able to reproduce the benefits if the liability was transferred to a personal pension, with competitive expenses.

With this in mind, and after consultation with the government with regards to what would be a reasonable basis, some assumptions were set. These are familiar and so are not repeated here. However, much has happened since then, for example:

  • expectations for price inflation have fallen;
  • evidence has come out to suggest that pensioners are living longer than previously anticipated;
  • real gilt yields have fallen;
  • the growth in UK equity dividends has been significantly lower than assumed;
  • the ability for UK pension funds to be able to reclaim tax credits on UK equity dividends was removed;
  • serious question marks have been raised over the methodology underlying the MFR.

Some minor tinkering has been made with the MFR to recognise these changes, but does the MFR in its current format still meets its original objectives? What would MFR look like now if it were to be set from scratch so as still to meet the original objectives? When MFR was originally set, it was based on gilt yields and long-term equity dividend yields close to what they were at that time and equity returns (net of expenses) were assumed to be 1% above gilt yields. If a similar approach were to be adopted now, the following basis could be derived in January 2003:

Gilt returns: 4.6% pa
Equity returns: 5.6% pa
Price inflation: 2.3% pa
Mortality: PxA92 (year of use) rated up 1 year
No market value adjustments

Figure 1 shows how the liabilities would compare under this ‘new MFR’ with the current MFR for non-pensioners with LPI pension increases. The ‘new MFR’ would produce liabilities about 5% higher than under current MFR at the older ages and around three times higher at younger ages. Furthermore, if the current MFR were to be compared with the cost of actually buying out the benefits with an insurance company, it would probably only secure around 15% of the benefits for a younger member.

The situation looks far worse for a scheme with fixed increases in deferment and in payment. Here the ‘new MFR’ would produce liabilities around eight times the liabilities under the current MFR at the younger ages. This is shown graphically in figure 2 ( see PDF).

In order for a ‘new MFR’ to produce similar liabilities to the current MFR, it would be necessary, for LPI-type benefits, to assume that equities would outperform gilts by over 4% pa in current market conditions. For fixed-type benefits, the necessary assumption for the outperformance of equities over gilts would increase to over 6% pa. This will depend on what allowance is made for mortality improvements and on whether any allowance is made for investment returns in excess of gilt returns after retirement.

The purpose of this article is not to recommend a ‘new MFR’ or to offer any comment on whether the current MFR is too ‘weak’. Some fundamental questions as to what is a reasonable security level for final salary benefits and how this should be met need to be answered first. Neither is the purpose to consider what would be an appropriate methodology for a ‘new MFR’. However, it does show that for many schemes the current MFR may not currently be suitable for transfer values or funding.

For a full version of this article see the PDF of the professional news