Examination of Witnesses (Questions 80-99)
MR DICK
SAUNDERS, MR
STEPHEN HADDRILL
AND MS
CHRISTINE FARNISH
25 APRIL 2006
Q80 Mr Todd: You have heard Ned Cazalet
putting a large critique of the competence of your industry to
predict accurately what is viable and what is not. He may have
differed from that and thought he was characterising you entirely
inaccurately, but that seemed to be what he was saying.
Mr Haddrill: I think it is important
to recognise that we are talking about a different kind of system
from the current one. I think Ned was talking very much about
the past rather than the future NPSS.
Q81 Mr Todd: I think he chose to
predict the future based on the past.
Mr Haddrill: Yes, he chose to,
and I would predict it slightly differently.
Q82 Mr Todd: There is just one last
point. You touched on the difficulties with the PAYE system and
the possibilities. You have heard my previous questioning, which
broadly produced an "I am sure we can deal with that"
answer. Is it your view that with a bit more rigorous thought
we can deal with the interface with the PAYE system?
Mr Saunders: I think it will need
a bit more than rigorous thought. It would be quite a significant
project. I would not claim to have sufficient expertise in the
administration of the tax system to be able to give you a good
answer there. One point I would make on administration, which
I think was touched on in the earlier session, is that one of
the problems that has been mooted about the tax system is that
you cannot reconcile until over a year later because it is only
when the tax returns go in, the P60s, that you are able to attribute
the tax payments to the individual. The question therefore arises
that if you are using that system to collect contributions under
NPSS, do you then have a problem because you cannot invest the
contributions? So long as you have a unitised default fund where
people own a number of units in that fund, then what you do is
that you park that money in a nominee account which is investing
in those units; then you are able to allocate those units at a
later date. The individual has not lost out on any investment
gain over that period.
Ms Farnish: Occupational schemes
already have the systems to hand money over to someone who is
investing in the fund on behalf of members, so I do not think
we envisage a major problem in this area.
Q83 Chairman: Who would bear the
risk of administrative cost or capital expenditure over-runs under
each of your models? Would it be the savers, the Government, or
individual firms?
Mr Haddrill: I think you have
identified where it has to fall. For the API model, it would fall
partly on the shareholder and partly on the customer, depending
upon what a regulator would allow. With the other models, it may
fall between the customer and the taxpayer, depending upon what
the Regulator and the Government decide. We are talking a lot
about the independence of the NPSS from the taxpayer and from
Government. Anything that Government sets up it is very hard for
it to walk away from at the end of the day if it goes wrong. That
is where we feel that the risk to the taxpayer comes.
Mr Saunders: In the context of
the NPSS, the glib answer is that that is entirely a contractual
matter. There will be a contract between the NPSS, which is the
commissioner of the service, and the service providers, which
would be admin providers and the fund managers, and there will
be a number of different contracts. Those contracts will go into
great detail. One of the things that they will have to address
is who bears the risk in the event of various eventualities arising.
Certainly, if I was negotiating that contract on behalf of the
NPSS, I would regard it as my duty to try to ensure that as many
of those risks rested with suppliers as possible.
Ms Farnish: Uniquely, the particular
model we have put forward does not require the development of
a major new system to act as a clearing house and a major piece
of infrastructure that does not exist at the moment. In terms
of new capital required to run a system of very large, trust-based
schemes, the capital would be very modest indeed. We think you
could grow a number of existing large, multi-employer schemes
to become these institutions very easily and very cheaply indeed.
Q84 Kerry McCarthy: We turn to the
question of consumer decision-making and consumer choice, which
we talked about earlier. I start with Christine Farnish. Under
your model with the process of the super trust, that means that
asset allocation decisions are taken by the trust rather than
by consumers. Presumably that is predicated on the notion that
consumers are not well informed enough to make that choice. Is
that the case and do you think that the disbenefit to consumers
from removing that choice will be offset by lower costs?
Ms Farnish: All the evidence shows
that the vast majority of consumers are not able to make those
sorts of asset allocation decisions. I think in virtually all
countries in the world that have defined contribution pension
savings systems, around 90% at least of consumers end up not able
to make a choice and so they default into whatever the default
fund offers, because they just cannot choose. Those that do make
a choice often end up choosing something which really is not in
their interests: for example, young people in their 20s choosing
an incredibly safe and cautious and low-risk fund which really
will not grow their pension savings over their lifetime, or, perhaps
just as bad, people choosing something that looks very racy, that
is high risk and really could deliver them a nasty shock before
they come to retire. Choice in individual investments is not,
we think, desirable as a policy objective if we are building a
new pension system for the mass market. That does not mean to
say that you could not build choice into super trust asset choice.
Obviously there is more cost associated with doing that because
you have to help consumers to make that choice. Is it a desirable
policy objective? We think it is not for this stage in the development
of the UK market. We think it would be much better for most people,
in terms of the overall return they would get and the way in which
the investment risk would be managed for them, instead for them
to have a personal account in a fair share of a large, pooled
fund which would be invested in a diversified range of assets
by people who are professional investment managers and who would
be charged with a statutory duty to deliver good returns and manage
risk on behalf of those consumers. That does not mean that people
would not take responsibility. They would still be taking responsibility:
(a) for being enrolled into the scheme and for saving; (b) for
adding any top-up savings which they might want to make into the
scheme; and (c) obviously then for the annuitisation process at
the back end.
Q85 Kerry McCarthy: Are you able
to quantify the cost savings that come from that?
Ms Farnish: We can very easily
because there is already masses of evidence about the costs that
are incurred by large, multi-employer, trust government schemes.
It is not at all difficult to do this for around the 30 basis
points mark that Turner has envisaged. We have built in some extra
cost in our modelling because we have added more cost for people
who might move from super trust to super trust or people who might
have more help through call centres and the like with deciding
whether to remain opted in or indeed with the annuitisation process,
which I must say we have not mentioned yet in this debate but
which is a hugely important part of this new DC savings world.
Q86 Kerry McCarthy: Could I ask the
other two then: are you convinced that you have a choice based
model so that consumers can exercise choice?
Mr Saunders: I would agree with
everything Christine Farnish has said. All the evidence from the
range, be they occupation, DC, from Sweden or wherever, is that
90% plus of people go for the default option. In designing the
scheme, it seems obvious to us that what you do is focus your
effort on defining a really good default model in the way that
Christine has described it. You may well want to offer some choice
around the edge, but recognise that that is where the money is
going to go; 92% in Sweden go for the default option.
Q87 Kerry McCarthy: Although your
scheme is based on giving the consumers choice, you are assuming
they are not going to take it?
Mr Saunders: Our scheme is based
on starting with the default model, and then deciding how much
extra choice you want to give around the edges. Somebody mentioned
earlier that there are other options you might want to offer.
Our suggestion would be to set up a really good, independent,
professional board to run the NPSS and let them make those decisions.
I think Christine would say you should do the same with the trustees.
Mr Haddrill: I think I largely
agree. The Government says that what it would like to promote
is some sense of personal responsibility for people over their
finances. I do not think you want a model without any choice in
it. You have to recognise that where we are now most people will
go towards a default model and that will make sense because they
do not have the capability to make the right choices. For those
people who feel they can, then some choice at the edges, as Dick
Saunders was saying, is reasonable. We talk about a default model.
We have to be a bit careful about that. If we end up with one-size-fits-all,
then we will not be serving the whole range of people because
a young person ought to be much more heavily invested in equities,
even through a default model, than a much older person. We probably
need to tailor, as it is called, lifestyle even the default model
against people's age and circumstances.
Q88 Kerry McCarthy: Can you elaborate
on what you think the risk and return characteristics of the default
model should be?
Mr Saunders: Stephen has rightly
drawn attention to the lifestyle products that are now in the
market, and this is quite common now within DC schemes: that where
somebody is young, you put them into a very aggressive, high risk,
high return fund, because even if it loses money at that point,
it does not matter because they are not retiring until 40 years'
time. As you get older, so you need to adjust that asset allocation.
Asset allocation is the key here. In the investment world, techniques
do evolve over time. I would not want to specify in the legislation
exactly how you are going to do this. You need to give this profession
board I have been talking about the flexibility to decide, and
then they would come in front of a committee like this and explain
what they are doing and why. Broadly speaking, you would need
to give the boardand I think this would be a function for the
Secretary of Statesome overall remit against which it has to steer.
The one that we suggest in our paper is that the investments should
aim at least to match growth of average earnings over the lifetime
of the investment for each individual investor. That is where
that lifestyling point comes in. If you give them an overall remit
like that, a star to steer by, then you can develop your detailed
investment strategy around delivering that objective.
Ms Farnish: There is a little
difference here perhaps with a lifestyle fund for an individual
consumer, which is like a personal pensions contract for them
where they have to say when they sign up when they think they
are going to retire. You need to think that many years ahead.
Your life may well change and the world may move on, but in order
for that life-styling to work, there needs to be some sort of
fixed point when you start moving out of equities into bonds.
That is the way the computer algorhythms work. One of the possible
advantages in having a big pooled fund where risk can be shared
between the members is that there ought to be a lot more flexibility
there for people to change their retirement planning aspirations
over time and for the fund as a whole to meet their requirements
and for them to get their share of that fund when they retire
rather than them just as an individual with their own little pot
having to plan it for themselves.
Q89 Kerry McCarthy: Can I ask the
other two why you have not chosen to go down this model?
Mr Saunders: What Christine has
described is exactly what I am suggesting.
Q90 Kerry McCarthy: Why have you
decided that it is worth retaining an element of choice over asset
allocation?
Mr Saunders: This is because some
people will want that choice. A minority of people will want to
have the ability to exercise choice, but we should recognise that
it will be a minority. Therefore, we should design the scheme
accordingly. Your starting point for the scheme is the default
fund.
Q91 Kerry McCarthy: It is an option
for the 8% to 10%?
Mr Saunders: It is an option for
the 5% to 10%, or whatever, who want to do a little bit around
the edges.
Mr Haddrill: The debate has very
much focused on driving down the costs. If we could get into an
area where we focus on driving up the net return, that is really
what people need at the end of the day, the highest possible net
return. I personally believe that a bit of competition around
the return on the assets as well as on reducing the cost is beneficial
so that the fund management does not go too far.
Q92 Angela Eagle: Does choice increase
cost because everybody has to advertise and have people to try
to sell business? Is it not the case that the more providers there
are, the more the costs of administering this scheme will rise?
Is that not the big trap that we have got into with retail insurance,
as we discussed in the last evidence session?
Mr Haddrill: If we are going to
get close to the kinds of figures that Lord Turner is proposing,
then there has to be some limitation on choice. I agree with that.
I would not say that that means that there should not be any.
In order to limit that choice, we talk in our paper about having
a fairly standard contract between the provider and the employer,
for example. We talk about selling direct rather than on a commission
basis. We talk about the need to regulate the level of marketing
expenditure. Yes, I think you are right in that.
Q93 Angela Eagle: So it would be
below the £30 billion that Ned was talking about earlier?
Mr Haddrill: I think we can probably
aim for a bit below that, yes.
Q94 Angela Eagle: Do you think that
the ordinary consumer scepticism about the insurance and life
assurance market, after years of mis-selling and the kind of market
model which is very self-serving and does not serve the consumer,
is misplaced? Why, therefore, given your track record, should
the Government even think about the ABI model?
Mr Haddrill: I think the scepticism
about some of the things that have happened in the past is right.
I think the industry has to take that criticism on the chin. The
question now is, looking forward: is it recovering trust? I think
that it is seeing quite large and high levels of investment coming
through. I think people recognise that it is paying out pensions
at very high levels, and it is committed to improving its customer
service. It has just signed up to a new scheme to do that. Equally,
of course, the FSA is now there standing alongside it. It is now
managed I think much better in terms of a capital prudential insolvency
regime. Therefore, consumers and the Government can have confidence
in it going forward.
Q95 Angela Eagle: Again, the structure
of the industry as at the moment only serves the top 10%. This
kind of structure, this kind of pension scheme, has to serve everybody,
including people that your industry has not bothered with in the
past because they cannot make a profit on them. Therefore, why
should consumers trust the same basic business model with competition,
advertising, distribution sales forces and all the rest to do
something that needs to be simple and as cheap as possible?
Mr Haddrill: That is because we
are not proposing the same model. We are proposing to take out
a long distribution chain and take out commission and sell direct
and move into a new world. A lot of what Ned Cazalet said was
right: this is an opportunity for the industry as well as an opportunity
for savers.
Q96 Angela Eagle: In both your models,
would you offer different prices or scheme charges to different
employers and, if you did, would large employers get it cheaper
or smaller employers get it more expensive?
Mr Haddrill: Not in ours, no;
it is standard.
Q97 Angela Eagle: So there would
be no differentiation at all between individual employer circumstances?
Ms Farnish: We see some differentiation.
We see benefits from having more than one institution out there.
The Turner model has one institution, which of course is very
simple and ought to be cheap, certainly cheap to begin with; whether
it is the cheapest answer in the long run is another question.
Single institutions usually get fairly sluggish and there is no-one
there to keep them on their toes. For a limited number of institutions,
like the sort of model we have put forward, we can see that there
would be some differentiation, for example in the quality of consumer
service and in the sort of help, advice and guidance given to
consumers about their retirement planning, to take one example.
That could be an area where one super trust did differentiate
itself from another. Another could differentiate itself on the
grounds of its administrative systems or its IT systems. So we
would hope that there would be good practice that would emerge,
that natural innovation would take place and these institutions
would learn from each other, as they would in any market, through
a very transparent, regulated process. We do not see a significant
difference emerging in the charges because the sorts of costs
that they would be able to operate would be regulated by, as we
are suggesting, the Pensions Regulator to make sure that all the
institutions there did operate at a certain level of cost.
Q98 Angela Eagle: Standard Life have
done some calculations on the Turner recommendations with 4% contributions
and revealed that, even on optimistic assumptions, this fund will
not break even until 2035. Are any of you worried about that?
Mr Saunders: We spend quite a
lot of time looking at the Turner costings. We have done a certain
amount of modelling of our own as well. The first point to make
about the Turner costings is that they are 20-year costings, so
that in effect it is suggesting that if you were to give a 20-year
contract to somebody to run the scheme, then you would expect
them to charge that 0.3% over that full 20-year period. If you
only give them a 10-year contract, it might well be that it is
slightly higher. This takes us back into Ned Cazalet's evidence
earlier. I completely agree with him that persistency is one of
the important factors in this, but only one. One point I would
make about Turner's costings is that on admin he is actually extremely
cautious if not pessimistic. We find it very difficult to see
how he gets to the £500 million he assumes for set-up costs
into a scheme. If you look at account admin, he talks about £25
per account. In Sweden, it is run at less than £5 per account.
In the TSP in the States, which you have heard about, it is run
at about $25 per account. One of the leading administrators in
the UK has estimated to us that once this thing gets up to scale,
they can run it at £10 per account or less. Paymaster, the
pensions administrator, has come out with a paper suggesting that
it can be done for £10 per account. Against all of that,
the Turner £25 per account does look quite cautious
Q99 Angela Eagle: Is anyone else
worried that this scheme, the fund, does not break even until
2035?
Mr Haddrill: I do worry about
it. It was one of the things that drove us to look at what we
could do it for. We felt that at least for the first 10 years,
the exiting infrastructure in the insurance industry would be
cheaper than the Turner model. I also worry about it because of
the risk of those numbers going up if Turner has got some of those
fundamental assumptions wrong. We do feel he is quite optimistic
on this point about persistency and participation in a non-compulsory
environment. Yes, I think it is worth worrying about. The other
worry is that the political environment on the Government's framework
will not necessarily stay the same. If you are looking to make
a return or for something to break even over 20 years, and someone
was asking what happens if the state system changes, then those
numbers can change again. I am not sure that we can be confident
that the state system will not change within the 20-year period.
| 
