Examination of Witnesses (Questions 1
to 19)
MR MICK
MCATEER
AND MR
NED CAZALET
25 APRIL 2006
Q1 Chairman: Good morning and welcome
to the first evidence session in the National Pension Savings
Scheme inquiry. Can you introduce yourselves for the shorthand
writer, please?
Mr McAteer: My name is Mick McAteer,
principal Policy Advisor at Which?.
Mr Cazalet: I am Ned Cazalet and
I, coincidentally, work for Cazalet Consulting, which advises
financial institutions on setting up life assurance companies
and pension funds.
Q2 Chairman: What are the main challenges
the Government faces in taking forward one of the models of the
NPSS, and what are the key decisions that have to be made?
Mr McAteer: It is our view that
the NPSS represents the best opportunity in a decade to design
second-tier pensions that act in the consumer's interest and pensions
that are designed from a consumer perspective. We used a number
of criteria to assess the different models from the IMA, the NAPF
and the ABI and to work out which would be the best option for
delivering the NPSS. We looked at competition aspects, cost and
affordability, choice, the risk to consumers, the levels of confidence
and trust in the various models, the regulation and legal security
in different models and the political risk and sustainability
of the different models. We concluded fairly strongly that the
NPSS and the IMA's model, which we think is the closest to the
NPSS, is by far and away the best model. That was followed by
the NAPF's model which came close to the NPSS system but had a
few drawbacks and a long way back was the ABI's proposals. We
simply cannot see how the retail model, as proposed by the ABI,
can meet those all-important cost considerations. Crucially, the
ABI proposals would not have the trust and confidence of consumers
and without that confidence and trust you will not have a sustainable
pensions system.
Mr Cazalet: I cannot comment on
the NAPF. Before I saw Lord Turner I spoke to some of his colleagues
and we looked at the numbers in Appendix F of the report and thought
let us put this into a model to see whether this would make any
sense for anyone to operate under any terms whatsoever. While
my firm said about five years ago that if you had a compulsory
pensions system and collected contributions via taxation or NI
you probably could run it for 0.2/0.3%, when we put the numbers
into the model to see whether it would be worth investing in a
model to run this on behalf of the Government, because someone
has got to run it, we found there were a number of problems. You
may have seen a document called Polly Put the Kettle On
which I think was circulated to you.
Q3 Chairman: You are the author.
Mr Cazalet: I am afraid I am the
author. On pages 33, 34 and 35 what we do is show some of the
outputs of that model. We assumed that people would save £100
or £130 a month or what have you and then we put it into
the model to see how much that would cost to run and we used Turner's
pricing. You will see on pages 34 and 35 the cumulative cashflow
charts and they mostly never get above zero, which means this
thing does not make any money. The problem we have with the Turner
model—and I suspect that they did not spend an awfully long
amount of time trying to model this, at least that is my understanding
of it—is that on a cost basis it does not work. What we
did in the model was we modelled persistency and we said how are
these people going to behave? If you look at the life assurance
system we have at the moment where people voluntarily sign up—okay,
they are persuaded, there is a salesman there, but they make the
decision at the end of the day to sign the form to say they are
going to start a pension—and if we take 100 personal pension
plans that are started today, in four years from now we could
expect less than half of those to be in force, that is the persistency
pattern. People lapse their policies for whatever reason. In our
document you will see we have shown the trends in persistency
using data collected by the FSA and using our own data. What we
could not understand is how in the Turner model, Appendix F, Turner's
colleagues had assumed this miraculous overnight transformation.
Given this is not a compulsory scheme, you might get trawled in
on the net on day one but there is no-one locking you into the
scheme, we had no satisfactory explanation as to why, given the
historic trend in terms of persistency, people were not continuing
with their contracts for whatever reason. Why does this suddenly
disappear overnight? Looking at this realistically, that is a
fundamental issue here. We say that if you have a compulsory scheme—and
there are huge problems with a compulsory scheme because of the
impact of means testing, the fact that you may compel people who
genuinely cannot afford to save and all sorts of other issues—and
if you do not take proper account of people's propensity not to
keep up the payments, then I am afraid that anyone that believes
Turner at a profit is deluding themselves. We did not just model
one set of inputs and outputs, we modelled it this way and that.
Frankly, if someone asked me to put my own personal money into
a vehicle to run this I would say "No way, Jose!". At
the end of the day if NPSS comes alive someone has got to run
it. It will be an infrastructure product. Someone is going to
have to set it up and someone is going to want a rate of return.
Just one last point on the cumulative cashflows we show for Turner
and you may want to compare those with the other cashflows we
show for the life industry model which are also abysmal. We did
allow a lower cost of capital for the NPSS on the basis there
would be some sort of Government backing, so we used a lower discount
rate, but what strikes me is will this thing work—leaving
aside Turner's great ambitions and I am very supportive of his
overall thinking—and is this a commercially viable venture?
I cannot see how that it is.
Q4 Chairman: We only wanted a short
inquiry on this, but it will be even shorter based on your evidence.
The whole thing is caput, is that what you are telling us?
Mr Cazalet: It is not for me to
put the match to it.
Q5 Chairman: Will this require Government
money to keep it going?
Mr Cazalet: I am aware of some
research that the ABI has done. I cannot validate this. I do not
know whether the numbers are right or wrong. The ABI's pensions
and savings team did trot off to Sweden and they looked at the
pricing of that model and at the cross-subsidy and so on and so
forth. The ABI's view was that that was also fundamentally loss
making. What we say is we have run the numbers, our numbers have
not been challenged, Adair Turner has looked at our numbers and
no-one has knocked at the door or phoned us up and said we are
speaking complete nonsense from the DWP and they have had plenty
of time to do so. These numbers have been in the public domain.
If I were on your Committee I would really be saying this is all
very well, but are we wasting our time having a hot air convention
if this thing cannot work on this pricing? What is really going
to happen to people who save in pensions? People stop saving.
They get pregnant, they lose their job, they get bored or they
need the money for something else. That has happened year after
year after year. If you do not compel them you are going to have
some lapse rate, some persistency problem. We have modelled this
using very optimistic persistency numbers compared to the current
situation and the numbers still look awful.
Mr McAteer: Ned is right if he
thinks that Lord Turner imagined the NPSS being run by retail
insurance companies. He is dead right, it is unsustainable in
that sense because the economics of access do not allow the retail
model to provide the NPSS on any charging terms that make sense
for consumers to save into. The key point is that Lord Turner
is not talking about the retail model, he is talking about a collective
pension scheme similar to the Federal Thrift Savings Plan which
uses bulk buying and economies of scale and competitive tendering,
and because you have a single account that means all the problems
of persistency and so on and transferring from account to account
are actually mitigated considerably. That is why, for example,
when you look at the Federal Thrift Savings Plan, what they do
is they collect the money centrally and they use competitive tendering
to allocate the management of the funds to the big institutional
fund managers on Wall Street. That is why the total operational
costs of the Federal Thrift Savings Plan are in the region of
0.35% a year and the investment management costs on their own
are in the region of 0.08% per year. No-one is talking about using
a retail product where firms have to compete for business.
Mr Cazalet: You are saying that
we modelled retail. We did not model retail at all. We have not
modelled or moulded the life assurance model in our workings for
the NPSS. We have paid no heed to that whatsoever. We took Turner's
own pricing from Appendix F.
Q6 Chairman: In Appendix F there
is a target of 0.3% for charges and I was going to ask you if
that were achievable, but obviously, in light of your previous
comments, the answer is no. At what level would this scheme be
achievable?
Mr Cazalet: I said earlier that
when we looked at stakeholder pensions some years ago, if you
had some centralised scheme but you then had excellent persistency—Yes,
we think the persistency should not be as bad under a Turner scheme
if you have one scheme and not a lot of competing providers because
there is quite clearly a lot of churn going on, so we do recognise
that persistency ought to be better, but it should not better
to the extent that would be required to make this profitable.
We can see how you could run a sort of scheme at 0.3% given an
economy the size of the UK and given a very good improvement in
persistency to the point where persistency does not become an
issue anymore. We did not back-test it to say where do we start
to break even on this, but my guess would be that within Turner
you are probably talking at allowing for some realistic persistency
numbers. There is no point pretending that persistency is just
going to disappear overnight, there is nothing that tells us that
is going to happen. So I think you are probably talking about
0.6% if you are going to take a realistic account of persistency.
At the end of the day, you do not want people to set a scheme
up and say let us presume that the world is suddenly going to
change and that nobody lapses. I take the point that some people
lapse because they move from one life insurance company to another,
but a lot of people lapse because they simply cannot keep the
payments up.
Q7 Chairman: So we have doubled Turner's
number without even trying.
Mr Cazalet: That is a rough number
to make it viable.
Mr McAteer: Lord Turner has said
that the 0.3% figure is a target and that is the whole point,
because over time in America what happened was the costs were
then bid down to about 0.35% a year. The basic point is that if
you have a collective scheme, a centrally run scheme where you
use competitive tendering rather than open market competition,
you deliver a model which is significantly cheaper than using
the retail model or the open market competition model. It may
not be 0.35% from the very outset, it may well be 0.6%, but the
point is, it will always be significantly cheaper than relying
on the open market competition model. That is the fundamental
point that people have finally grasped.
Q8 Chairman: Ned, you have made comments
about churnings in the industry and I think you have been supported
by Trevor Matthews of Standard Life who made a very full comment
about the basic business model of the industry being flawed. Under
the current insurance company business model are the substantial
payments to intermediaries, instead of promoting additional savings,
merely encouraging the recycling of existing pensions between
different providers?
Mr Cazalet: There is quite a lot
of evidence to support that. I do not know whether you are going
to have representatives of various insurance companies coming
here. We have been told things in confidence so I cannot give
you insights into individual companies. There is a great deal
of concern about the business model that these guys themselves
are operating because when somebody does build up a sizeable pension
pot the temptation is there, one way or the other, for this stuff
to suddenly be exited from "XYZ Life" and then re-brokered
with "ABC Life". You only need to look at the commission
terms that are attached to these things, on what we call the lump
sum business, the single premium business, which the industry
now refers to as transfer business, so this stuff is being transferred
around the houses. If you look at the commission terms in our
document on page 20, what we did is set out some real commission
levels applying at the back end of last year, but one of the things
we showed in the table was the commission claw back period. Why
do you think anybody would want to put a claw back period in there?
It varies from no years to three years to two years to four years.
This is part of the pricing. There is evidence to show that in
some cases some intermediaries have got a "cuckoo clock"
that comes to life three years down the line and says to the intermediary,
"It's time to review the case. You have made your commission.
You are not scot free. You can move your client"—without
necessarily being detrimental to the client—"from one
provider to another." This economics is lunatic. If you get
Trevor Matthews of Standard Life and others saying it is lunatic
then I suggest that this industry's business model is fundamentally
flawed and therefore it has got to come to a stop at some point.
I guess one of the big issues for the insurance company model
at the moment is that people have talked about the client and
the customer, but what they are doing is they are competing for
distribution, they are competing to get the intermediary on their
side. Hence this stuff may look mad and it is mad, but I think
people have got fixated in this desperate struggle to get new
business models, to kill everybody else, to be the king of the
castle and let us hope that it sorts itself out down the track,
but this is not economic, it is not sustainable. It is a bit of
an odd cartel. Normally the cartels are there to gouge profits
out of the consumers and keep everybody else away. This keeps
everybody else at bay because nobody in their right mind would
come into this industry from scratch and set this up. You would
not invest your own money in a company that I set up tomorrow
and said let us call "McFall Life" or "The House
of Commons Life Assurance Company". You would think I was
a lunatic. Companies are engaged in this struggle for survival
and there is a big emphasis placed on the new business numbers,
particularly for the listed companies.
Q9 Mr Todd: I want to explore the
administrative arrangements. Mick, you have poured water on the
idea of using the existing administrative systems of insurance
companies to provide the backbone of the new products. Why do
you reach that conclusion?
Mr McAteer: The conclusion we
came to was that certain of the vested interests are trying to
play up this idea that it makes economic sense to build on the
existing infrastructure and therefore it would be a cheaper administrative
set of arrangements. We actually had a look at this and we broke
down the whole collection of functions, administration, reconciliation
and verification, into its different component parts and we thought
there were several administration elements and so on. You may
well be able to use existing insurance company infrastructure
to collect contributions and so on, but at the end of the day
you are still going to need a verification and reconciliation
function and that can only ever be run by some kind of centralised
collection and verification agency anyway. If you have a system
that requires employers and employees to contribute and you are
using taxpayers' money then there will have to be some sort of
centralised function to ensure that that system is running smoothly
and that employers and employees are not breaching their statutory
obligations within a National Pension Savings Scheme. We found
it very hard to see where the existing infrastructure would have
significant advantages.
Q10 Mr Todd: You recognise there
is a balance of risk to be attached to this as well and that setting
up any centralised administrative structure to run this will involve
a significant project being put together in which failure is possible.
Mr McAteer: Yes. I have not met
anybody who is arguing for the Government to run the administration
of the NPSS.
Q11 Mr Todd: You certainly will not
find me arguing for it.
Mr McAteer: It is one of these
"straw men" I think the industry is trying to put up
to divert attention away from the flaws in their own business.
Q12 Mr Todd: Your model would be
a third party carried out on a competitive tendering basis.
Mr McAteer: Exactly. If you look
at National Savings, National Savings outsources the administration
of all its savings accounts to Siemens, the big IT supplier. If
you go to America and you look at the Federal Thrift Savings Plan,
they outsource the administration of that savings plan to a third
party as well on a competitive tendering basis. No-one is arguing
for the Government to run the administration of this scheme anyway.
The fundamental point is that, whichever model you adopt for administration,
you cannot get away from the need for some kind of centralised
clearing-house. It all has to come back to that eventually.
Q13 Mr Todd: Ned, do you think the
insurance companies have anything to offer in terms of administrative
competence to build on in this?
Mr Cazalet: I guess for the insurance
companies there is a range of administrative capability. Some
people are struggling with the business they are writing today
and they have poor systems or complex systems or out-of-date systems
and it is a function of running long-term insurance in part because
you write a contract and you may be on the books for 40 years.
You can see why people have written contracts on old, clunky,
green screen machines dating back practically to the Ark. I think
there is a range of outputs. There are some good admin systems
run by insurance companies. Some insurance companies do outsource
their admin to third party administrators. The thing that strikes
me about this admin bit is that, first of all, it has got to be
run by somebody and it has got to be run on commercial terms.
Does the whole thing stack up? The other thing as far as the insurance
companies are concerned is what are you administering? The life
assurance model is changing in front of us with the so-called
"open architecture". If you cannot do it already, you
will be able to go soon to the likes of Legal & General and
Standard Life and have access to 500, 600 or 700 funds. Given
that that sort of concept is already in existence and given that
people are investing their money and they want a gateway, I do
not see how you would have seven or eight insurance companies
competing for this at any price because they cannot afford to
run the business as it is at the moment.
Q14 Mr Todd: You are broadly favouring
some centralised model of administration, are you not?
Mr Cazalet: If you are going to
have a centralised system then have a centralised model. It may
well be that you have a competition amongst insurance companies
who do have some skills or you outsource it, but I cannot see
that you are going to have six insurance companies all active
on day one—
Q15 Mr Todd: So you are not in favour
of the "straw man" that is being talked about of a government
agency carrying out this function?
Mr Cazalet: It may be a government
agency, but the government is going to devolve this and outsource
this activity to somebody or someone.
Q16 Mr Todd: There has been speculation
that this is a government task and everyone has waved goodbye
to failed IT projects. Let me say from my experience in the private
sector that failed IT projects are not unique to the public sector.
Is this something that people should be concerned about?
Mr McAteer: There has been lots
of speculation in the press prompted by vested interests that
the Government wants to run a quango. Believe me, I have not met
anybody who is seriously proposing that the Government should
run the administration of the National Pension Savings Scheme.
It is a straw man created by people trying to divert attention
away from the flaws in their own business model.
Q17 Mr Todd: Surely the best approach
would be to design an administrative model in co-operation with
the sellers of any products that will pay into this and then contract
it out and replace it with a competent third party by competitive
tender.
Mr McAteer: I think competitive
tendering would be the best way to do it, but it depends on the
structure of the system itself.
Q18 Mr Todd: I think you have made
some reference to the difficulties of dealing with PAYE systems.
Could you expand on that slightly?
Mr McAteer: Some people have raised
concerns that if you use the existing PAYE system then there might
be some time delays in allocating money to people's accounts,
but I think the Investment Management Association has dealt with
that very well in their submission. They advocate a system of
nominee accounts which actually would mean that people would not
lose out on any of the contributions. I think that is technically
feasible.
Q19 Mr Newmark: I want to turn to
asset class and fund choice. Given that some analysts believe
that asset allocation decisions can explain up to 90% of the returns
enjoyed by savers, are you convinced that consumers possess the
necessary expertise to make asset allocation decisions themselves?
Mr Cazalet: No.
Mr McAteer: It all depends entirely
on the range of choices that are put in front of them. I am quite
confident that if you had a very, very restrictive range of choices
consisting of a default fund, a guaranteed fund and one or two
basic other funds then ordinary consumers could spot the difference
between a risky fund, investing in equities and a safer fund which
invests in bonds and so on. You have to provide a decent "traffic
light" system.
| 
