[Skip to content]

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

Changes to the MFR

been a hot topic over the past few months.
In September, the Pensions Board of the
Faculty and Institute of Actuaries (F&IA)
released its review of the MFR, commissioned by the
Department of Social Security (DSS). At the same time
the Treasury published a consultation paper that was
prepared by the DSS. This sparked a flurry of reactions
from participants in the pensions industry.
The DSS and others raised concerns that gilt and
bond markets would not be able to cope with the
increased demand resulting from the suggested
changes. This article analyses whether these concerns
are justified, and comments on potential alternatives
to the MFR.
The MFR review
The most significant change suggested in the actuarial
review is to base the liability valuation entirely on
bond yields rather than a combination
of gilt and equity
yields. At the same time, the
review suggests lengthening
the deficit correction periods
(from one to three years to
bring the funding up to 90%,
and from five to ten years to
achieve full funding).
For illustration, table 1 summarises
the average asset allocation
of UK pension funds
and the liability matching
portfolios under the current
and suggested revised MFR
bases for a typical scheme.
(The term ‘matching portfolio’
used in this article is a
notional concept, defined by
the MFR valuation methodology.
It generally differs from
matching portfolios identified for asset/liability management)
The typical scheme is based on the average
scheme profile described in the review (25% pensioner,
75% non-pensioner liabilities) with less than
£100m in pensioner liabilities and a small percentage
of guaranteed minimum pension (GMP) pensions.
The table shows that the revision increases the mismatch
between pension fund asset allocation and the
MFR liability-matching portfolio (the overweight
position in UK equities and the underweight position
in bonds both increase. (The actual positions will be
scheme-specific, but the figures above are a good illustration
of the position of most funds. The direction of
the changes is the same for all schemes.)
Some simple calculations show that the capital market
implications would be major if all pension funds
fully adopt the liability-matching portfolio. Based on
table 1 and an estimated size of UK pension fund
assets of £800bn, the net result would be a sale of
about £600bn equities (£400bn UK and £200bn overseas)
and a purchase of roughly £450bn in indexlinked
gilts and £250bn in corporate bonds.
The size of the UK equity markets is about £1.3 trillion
and the demand for index-linked and corporate
bonds would be larger than the total market sizes
(£70bn and £190bn respectively).
It should be noted that the corporate bond-backed
liabilities are not actually matched by a composite of
available corporate bonds, because the inflation
assumption in the valuation basis would be based on
the difference between fixed-interest and index-linked
yields. Consequently, the matching asset would be an
index-linked corporate bond portfolio. However, the
index-linked corporate bond market is growing, but is
still very small less than £10bn so that it is virtually
impossible to invest in the liability matching
Capital market implications
The above calculations are too simplistic to determine
the capital market implications. Not all pension funds
would move all the way to the matching portfolio at
once, for several reasons:
°ª MFR is only an issue if the funding level is close to
100%. The F&IA review reports that the MFR funding
level of the median fund is around 120% and
about 15% of schemes are below 100% funded. Consequently,
the investment strategy of a majority of
the funds need not be significantly affected by MFR.
°ª Pension funds do not adjust their strategic allocations
frequently or radically. This is illustrated by the
market reactions around 14 September 2000 (when
the review was published). If this had induced pension
funds to buy gilts and corporate bonds, one
would have expected gilt yields to fall and credit
spreads to tighten. Figure 1 shows the ten-year gilt
yield (Treasury 5.75% 7 Dec 2009) and the swap
spread (difference between the ten-year swap rate
and the ten-year gilt yield) at close on the days surrounding
the publication. This shows that after publication
the gilt yield fell slightly and the swap
spread increased marginally. The fall of the gilt yield
is in line with increased demand for gilts, but is
small and not abnormal compared to ordinary market
volatility. The increase in swap spreads is contrary
to expectations but, again, small compared to
usual market movements. Strictly speaking, this is
an indication of little reaction, but not definite proof, as pension funds could have anticipated the
contents of the MFR review prior to publication.
In general, it is difficult to assess the impact of asset
allocation changes on bond yields and equity market
levels, because this involves predicting the behaviour
and interaction of all market participants, ie pension
funds, other investors, and equity and debt issuers.
However, it is likely that the suggested changes in
the MFR methodology, together with increasing
maturity and concerns about the size of the risk position
pension funds are taking, would drive pension
funds out of equities and into bonds, to an extent.
The size of the moves is unclear, but a large asset allocation
shift is not inconceivable over time, as pension
funds tend to follow each other to some extent. Consequently,
an initiating shift by several funds could
result in a snowball effect.
Our simple initial calculations show that the capital
market impact is likely to be larger in bond than in
equity markets, because of the size of the markets. In
addition, the UK bond markets are not very liquid
compared to, for example, the euro market. The euro
market is about three times the size of the sterling
market, but the liquidity is about 20 times as high.
The limited sterling liquidity is a result of a small
number of large players pursuing similar investment
objectives and following buy-and-hold strategies.
In summary, the DSS and the Treasury are right in
voicing concerns about the impact of the suggested
MFR revision on gilt and bond markets. Possible
adverse market impact is often inappropriately used
in the pension fund industry as an argument against
change. However, the aggregate size of pension funds,
combined with the limited liquidity of the UK bond
markets, suggests that any significant move into
bonds over a short period of time would put downward
pressure on yields.
The F&IA review tries to mitigate the market impact
by lengthening the deficit correction period and basing
the liability discount rate on a broad range of gilts
and bonds. The first measure reduces the volatility of
cash injections on the downside, but the second measure
bears little significance. The long duration of pension
liabilities means that the matching portfolio will
still consist of long bonds.
In a wider perspective, the impact of a switch from
equities into bonds would be neutral if companies
simultaneously buy back equities and issue bonds.
However, it is unlikely that in practice such a transformation
of UK plc would be co-ordinated with a
change in pension fund investment needs.
Another mitigating factor is that the UK financial
markets are integrated to some extent in the world
markets. Consequently, some of the impact on the UK
bond market will be dampened by interaction with
global bond markets.
Alternatives to MFR
The suggested MFR revisions have also been criticised
for increasing the trend from defined benefit to
defined contribution pension provision. The reduced
funding flexibility and the expected higher contributions
resulting from lower equity allocations would
induce companies to move away from defined benefit
However, the critics should realise that the MFR is
designed to protect scheme members. By definition,
member protection and funding flexibility are conflicting
objectives. Consequently,
it is not
possible to design a
funding standard
that increases benefit
security and
maintains full funding
discretion for the
An alternative is to
introduce a central
discontinuance fund
(CDF) for financing
shortfalls in case of
wind-ups. This would also take some pressure off bond
markets. However, if the premiums into the CDF are
based on a funding test (in line with the Pension Benefits
Guaranty Corporation (PBGC) in the US), underfunding
still has direct cash consequences.
Paul Myners suggested abolishing MFR altogether
and replacing it with a combination of increased
reporting transparency and fraud protection. His suggested
reporting format is basically an executive summary
of an actuarial valuation and would leave
considerable flexibility in setting valuation assumptions
and contribution rates. This would offer limited
protection to members, as they would have to interpret
and challenge the reported results, including the
subtle relation between valuation assumptions and
funding position.
A viable and more objective alternative to Myners’s
suggestion would be to use the suggested revised MFR
basis for reporting the funding position annually, but
remove the deficit correction instructions. This would
maintain funding and investment flexibility (subject
to the scrutiny of annual disclosure) and increase
member protection by disclosing the level of liability
cover. Again, this does not maximise flexibility and
member protection at the same time, but it seems a
workable alternative that suits both objectives.
The impact
The impact of the suggested MFR revisions on capital
markets is difficult to assess, but it is likely that it will
increase the pressure on bond markets as pension
funds move from equities to bonds.
A workable alternative would be to implement the
suggested MFR revisions, but to remove the deficit
correction scheme. This leaves members at risk, but at
least they would have a single standard of reporting
and better information to press trustees to negotiate
security levels on their behalf.