Polar

Polarisation and
Financial Services
Intermediary Regulation
A review for the

Financial Services Authority
Report to the
Financial Services Authority
by London Economics
July 2000
Polarisation and Financial Services
Intermediary Regulation
A review for the Financial Services Authority
London Economics
June 2000
© Copyright London Economics. No part of this document may be used
or reproduced without London EconomicsÕ express permission in
writing.
Contents Page
1 Summary 1
2 Introduction and outline 4
3 The current regulatory environment 6

3.1 The introduction of polarisation 6
3.2 Description of the polarisation regime 7
3.3 The DGFTÕs 1999 report 12
3.4 Regulatory approaches in other countries 13
4 The market for polarised packaged products 15

4.1 Market shares of different distribution channels 15
4.2 Price trends 19
4.3 The provision of advice 21
4.4 Cost of advice 22
5 The economic effects of polarisation 25

5.1 Consumer detriment 25
5.2 Competition effects 40
5.3 Scenarios to be investigated 47
6 Technology and regulation 49

6.1 Technological change 49
6.2 Regulatory change 52
6.3 Conclusions 55
7 Options for change 56

7.1 Defining and assessing the scenarios 56
7.2 Scenario 1: Redefinition of polarisation for all products. 57
7.3 Scenario 2: The DGFT recommendation 69
7.4 Scenario 3: Minimum standards exemption 73
7.5 Scenario 4: Retaining the essentials of polarisation 75
7.6 Variant A: Panel/commission separation within IFA firms 77
7.7 Variant B: Only fee-based advice can be ÒindependentÓ 78
8 Cost-benefit analysis 83
London Economics
June 2000 i
Contents Page
8.1 CBA summary tables 84
9 Appendix: Detailed CBA 86

9.1 Scenario 1: Redefinition polarisation for all products 86
9.2 Scenario 2: The DGFT proposal 90
9.3 Scenario 3: Minimum standards exemption 92
9.4 Scenario 4: Retain the essentials of polarisation 95
London Economics
June 2000 ii
Tables & Figures Page
Table 1: The changing reduction in yield (RIY) on life and

pension products, 1995-1999 20
Table 2: The extent of commission rebating (percentage of new
regular premium personal pension business written) 24
Table 3: Market concentration (1993-1998) 43
Table 4: Concentration in gross unit trust sales 44
Table 5: Concentration in life and pension products in the tied
channel 46
Table 6: Remuneration of advisers 64
Table 7: Percentage of proprietary and non-proprietary funds
sold by the top five firms in the USA in 1998 67
Table 8: Significant incremental costs 84
Table 9: Significant incremental benefits 85
Figure 1: Distribution of life and pensions new business (APE),

£ millions, 1989-1999 16
Figure 2: Distribution of unit trusts and ISAs gross sales (APE), £
millions, 1994-1999 17
Figure 3: Breakdown of the sale of personal pensions by main
distribution channels, 1993-1999 18
Figure 4: Breakdown of the sale of life products by main
distribution channels, 1993-1999 18
Figure 5: The changing reduction in yield for personal pensions,
1992-1999, (25-year unit-linked personal pensions, £200
per month contribution) 21
Figure 6: Average five year cost of advice via different
distribution channels (£) 23
Figure 7: Reasons given for not buying a product (percentage of
those who seriously considered buying but did not
buy) 27
Figure 8: Barriers to buying financial products for inactive
consumers (percentage definitely agreeing or tending
to agree) 27
Figure 9: Demand for advice by distribution channel 29
London Economics
June 2000 iii
Tables & Figures Page
Figure 10: Extent of understanding of polarisation status by

recent purchasers of financial products, 1998 30
Figure 11: Consumers perception of the independence of their
financial adviser (by investment product) 31
Figure 12: Asked if Òa bank is the best place to get independent
financial adviceÓ 32
Figure 13: Proportion of consumers making initial contact (by
distribution channel) 34
Figure 14: Proportion of consumers making initial contact (by
product) 35
Figure 15: Proportion of consumers making the initial contact
the first time they used a distribution channel versus
the most recent time 36
Figure 16: Extent of shopping around for advice on savings and
investment products, 1996 - 1998 37
Figure 17: Did consumers using one channel consider other
channels? 38
Figure 18: The influence of advisers 39
Figure 19: Preferred means of obtaining information on financial
products 50
Figure 20: How would it affect your likelihood of using an IFA,
if they chose from a narrow range of companies? 60
Figure 21: How would it affect your likelihood of using a tied
agent, if they offered the products of several
providers? 61
Figure 22 Costs as a percentage of new business, 1997 63
Figure 23: Demand for advice, by product 70
Figure 24: Response of IFA customers, if tied providers were to
offer multiple unit trust products 71
Figure 25: Were you offered a choice of fees or commission? 79
Figure 26: How much would you be willing to pay for advice? 80
London Economics
June 2000 iv
Section 1 Summary
1 Summary
This report provides a review of the polarisation rules and an analysis of

alternative regimes for the Financial Services Authority (FSA). This involved
interviews with providers and advisers, consumer research, extensive
background research on market trends, and detailed consideration of the
underlying economic issues, to build up a comprehensive body of evidence
on the effects of the regime and possible changes to it. It examines four
possible changes to the rules using cost-benefit analysis. The scenarios mainly
emerge from the underlying economic analysis, and are as follows:
• Redefinition of polarisation for all products. In this scenario, firms

could choose to be independent, tied, or multi-tied, with clear
disclosure of both status and interest.
• The Director General of Fair TradingÕs (DGFT) recommendation.
In this scenario, collective investment schemes are removed from the
ambit of polarisation.
• Minimum standards exemption. In this scenario, simplified
packaged products (CAT ISAs and stakeholder pensions) would be
removed from the ambit of polarisation.
• Retaining the essentials of polarisation. In this scenario, providers

operating tied distribution are allowed to introduce the products of
other providers to fill gaps in their product range.
In addition, we considered two variants, namely the separation of product

panel selection from the negotiation of commission, and differentiating
between advisers remunerated on a fee basis (we call these ÒIndependent Fee-
based AdvisersÓ) and those remunerated by commission (which we call
ÒCommission BrokersÓ).
Our main findings are as follows. The polarisation rules appear to have some
anti-competitive effects, largely by blunting competition in the tied channel.
Redefining the rules would therefore bring some economic benefits,
particularly for consumers using this channel.
At the same time, the rules also appear to have had some effect in reducing
consumer detriment, by clarifying the status of advisers in the eyes of the
public. If the rules are to be changed, this should be done in such a way as to
maintain clarity about who is independent and who is not. Otherwise, there
is scope for consumer detriment to increase.
The key analytical consideration is therefore whether polarisation can be

changed in such a way as to have little impact on the size and independence
of the IFA channel. The evidence suggests that most IFA customers value
independence highly and would be unlikely to switch to a multi-tied channel
if this became available. The evidence also suggests that IFAs are in a
somewhat distinct market segment from tied agents, and this is reinforced by
1
London Economics
July 2000
Section 1 Summary
IFAsÕ different contractual relationship with their client. We conclude

therefore that the IFA channel would not diminish significantly if multi-tied
agents came into the market. However the evidence is not conclusive and
this is an area of risk.
The results of our cost-benefit analysis of the scenarios are as follows:
p Redefinition of polarisation for all products. Although such a change

would bring both direct costs to the FSA and compliance costs for
regulated firms, there would be worthwhile competition benefits, with
increases in product volumes and reductions in prices for consumers.
Meanwhile, there is unlikely to be much increase in consumer detriment.
In the IFA channel, this is because independent status remains protected,
and few current independent advisers would choose to abandon this
status. In the tied channel, other regulatory protections (the Òas good asÓ
provision) remain in place, and the importance to providers of protecting
their brands is likely to prevent firms from buying-in poor quality
products. There may be benefits in the bancassurance channel, where
there is substantial potential for falls in consumer charges to reflect the
low expense levels.
p The DGFT recommendation. This scenario may reduce the effectiveness

of competition, and it also has the potential to increase consumer
detriment for collective investment products, since it allows advisers to
use different statuses whilst advising on investment products that are
close substitutes.
p Minimum standards exemption. This scenario is less detrimental than

the DGFT recommendation. There is less scope for consumer detriment,
since the products removed from the ambit of polarisation are subject to
minimum standards. It would have little or no adverse effect on
competition, since it reinforces through status restrictions a segmentation
of the product market that is already being enforced through product
regulation. However, it would also bring few benefits because product
regulation only permits limited gains from competition.
p Retaining the essentials of polarisation. This scenario offers some gains

from enhanced competition between providers in the tied channel.
However, it offers no gain from competition within the product ranges of
tied providers. There could be some impact on the IFA channel but it is
likely to be minor. Overall, it is little different in economic terms from the
status quo.
2
London Economics
July 2000
Section 1 Summary
The two variants also have limited market impact. The proposal to separate
panel selection from commission negotiation, for IFA panels, would largely
formalise practices that are already in place, and would have few benefits or
costs. The separation of fee-based IFAs from commission-based brokers,
however, would result in the great majority of IFAs becoming brokers, with
substantial regulatory costs and some possible adverse effects on consumer
detriment (as a small number of customers would be displaced from
currently independent to tied channels).
In our view, three of the core scenarios (redefinition, minimum standards

exemptions and retention) form possible options, with redefinition being the
most economically attractive.
Clearly however the FSA will have to consider other factors alongside cost-
benefit results in deciding which course of action to undertake. It may for
example wish to select elements from several of these scenarios, and Òmix
and matchÓ to create a preferred regulatory outcome.
3
London Economics
July 2000
Section 2 Introduction and outline
2 Introduction and outline
The regime whereby advisory roles with respect to packaged investment

products are polarised into independent intermediaries on the one hand, or
tied agents on the other, was introduced in 1987. In 1999, it was reviewed by
the Director General of Fair Trading (DGFT), who recommended certain
changes to the regime in the interests of competition. In February 2000, the
Financial Services Authority (FSA) commissioned research to investigate
these and other possible changes to the polarisation regime. Specifically, the
FSA asked for an assessment of how four possible policy scenarios would
compare to maintaining the current regime. 1
In consultation with the FSA, the policy scenarios were refined somewhat
during the course of the project, as it became clear that some of those
originally envisaged would either not constitute good regulation, or would
not constitute significant changes in the regime. The scenarios finally agreed
with the FSA and presented in this report are as follows:
p The redefinition of polarisation for all products, so that firms could

choose to be independent, tied, or multi-tied, with clear disclosure of both
status and interest.
p The Director General of Fair Trading (DGFT) recommendation that
collective investment schemes are removed from the ambit of
polarisation.
p Minimum standards exemption. A proposal where simplified packaged
products (CAT ISAs and stakeholder pensions) would be removed from
the ambit of polarisation.
p Retaining the essentials of polarisation, while allowing providers
operating tied distribution to introduce the products of other providers to
fill gaps in their product range.
In addition, we considered two variants, namely the separation of product

panel selection from the negotiation of commission, and differentiating
between advisers remunerated on a fee basis (we call these ÒIndependent Fee-
based AdvisersÓ) and those remunerated by commission (which we call
ÒCommission BrokersÓ).
In addition to the submissions made to the FSA, LE has drawn on a number

of other sources of information:
• Consumer research: although the market for advice has been the
subject of extensive consumer research, we collected further
information to understand better how different types of consumers
1 Following the TreasuryÕs decision on the regulation of advice on mortgages, analysis of the market for
advice on mortgages was outside the scope of this project.
4
London Economics
July 2000
Section 2 Introduction and outline
end up using different distribution channels. The survey was

undertaken by Continental Research in April 2000, and involved
around 1000 interviews with recent purchasers of packaged
investment products. All the figures sourced to Continental Research
in this report are based on the survey.
• Structured interviews: to understand how firms might behave in the
future we needed to understand why they behave as they do today
and how they got to the current competitive position. Over thirty
interviews were undertaken by London Economics during April and
May 2000, with providers, advisers, industry associations and
overseas regulators.
• Evidence based review: To give a firm basis for the review,
particularly in areas where industry views vary substantially, we
developed a comprehensive database of evidence. This throws light
on crucial concerns such as the success of the current regime for
competition, the level of consumer detriment and the effects of
changes in technology and regulation.
The rest of this report is structured as follows:
p The next section sets out the original motivation for introducing the
polarisation rules, the detail of these rules, a summary of the Director
General of Fair TradingÕs recent report and some other countriesÕ
experiences with intermediary regulation.
p Section 4 describes trends in the market for both packaged investment
products and the complementary demand for advice.
p Section 5 shows how the polarisation rules affect competition in the
packaged investment market and their effect on consumer behaviour and
the resulting level of consumer detriment.
p Section 6 covers changes in technology and regulation that must be taken
into consideration in the evaluation of market outcomes.
p Section 7 describes the scenarios in detail and the likely market outcomes.
This focuses not just on the changes to the polarisation regime but also on
the changes to other rules necessary to support these new rules.
p Section 8 presents the results from the cost benefit analysis of these
scenarios.
The details of the cost-benefit analysis are included in the appendix.
5
London Economics
July 2000
Section 3 The current regulatory environment
3 The current regulatory environment
The current regulations governing the provision of financial advice are

described in detail in the FSA Conduct of Business Sourcebook2 and other
regulatory material. This section summarises particular topics that are
relevant to this review of the polarisation regime:
• the market and regulatory context at the time polarisation was

introduced;
• the specific rules relating to polarisation;
• the DGFTÕs 1999 report on the polarisation regime; and
• how the UKÕs regime for investment advice compares to those in
other countries.
3.1 The introduction of polarisation
Before the Financial Services Act 1986 (the Act), there were three broad types
of agents providing investment advice:
• agents of insurance companies, who worked solely for that company;
• insurance brokers; and
• firms or companies paid through commission for introducing

business to a small number of providers. These were largely bank
managers, estate agents and solicitors.
Although not a part of the Act itself, polarisation was introduced as part of
the regulatory framework in 1987. The Act reflected the review published in
1985 by Professor Jim Gower to improve investor protection.
The Act put into place a comprehensive system of regulation, covering

persons performing any form of investment business giving specific advice. 3
Before 1987 there were no rules defining the meaning of independent advice
or training of advisers. This situation resulted in agents purporting to give
independent advice when in fact they were representing a single firm or a
small number of firms. It was partly in response to this issue that the
polarisation rules were imposed.
2 With particular reference to CP45A.
3 Specific advice is authorised investment advice which is tailored to meet the needs of an individual
investor. General advice covers information about classes of product such as pensions.
6
London Economics
July 2000
Some of the main arguments made at the time for the introduction of
polarisation were that it would lead to:
• less investor confusion about advisersÕ status;

• less investor risk through less risk of conflict of interest; and
• greater clarity of where responsibility lies for the advice.
The arguments against polarisation were that it would prevent, restrict or

distort competition, and included the idea that it would lead to:
• costly disruption of commercial arrangements, requiring large

amounts of investment; and
• a weakening of the independent intermediary sector, as higher
regulatory costs would be imposed and many independents would
exit the market.
The Director General of Fair Trading (DGFT) was required to examine the
potential impact of these rules before the Securities and Investment Board
(SIB) was able to implement them fully. In his 1987 report 4 he concluded that
several adverse effects on competition were likely to result from the
introduction of polarisation:
• a significant reduction in the numbers of people who were able to

give independent advice;
• an increase in the influence of tied advisers; and hence
• a reduction in information available to consumers on competing
products.
Despite these concerns, polarisation rules were introduced by the SIB in 1987
as part of the new regulatory regime for investment business. The other
regulatory bodies set polarisation rules at least as strong as those
implemented by the SIB.
3.2 Description of the polarisation regime
The polarisation rules require that:
Ò1. A firm which advises a private customer on packaged products must

either:
a) be a product company or its marketing group associate; or
b) do so as an independent intermediary.
4 Securities and Investments Board: A report by the Director General of Fair Trading to the Secretary of
State for Trade and Industry, March 1987.
7
London Economics
July 2000
2. A firm which is a product company or its marketing group associate must

not advise private customers to buy packaged products which are not
those of the marketing group.
3. A firm which acts as an independent intermediary in advising a private

customer on packaged products must act as an independent intermediary
whenever it advises private customers on packaged products in the
course of regulated business.Ó 5
Polarisation only applies to advice on packaged investment products. These

include:
• life policies;
• personal pensions; and
• collective investment schemes, investment trust savings schemes,

whether held in a PEP or ISA or otherwise.
3.2.1 Independent advice

Independent Financial Advisers6 (IFAs) are advisers who may not have any
tie with product providers. They are required to act on behalf of the investor
and Ð by surveying the market Ð advise them on the most suitable product.
An IFA can be the appointed representative of an IFA network. In this case,

the network is responsible for the training and competence requirements of
the IFA and for compliance with the rules.
IFAs need not provide advice on all product types and can specialise in
certain areas (e.g. annuities, pensions). But where the advice is provided, it
must be for a product which is as good as any other on the market. An IFA or
IFA network must take responsibility for the advice it offers.
Some banks and building societies have IFA arms, particularly to advise high
net worth clients. Since these banks and building societies also offer their
own products, there is a possibility that without further regulation, these IFA
arms would be biased towards offering in-house products. In order to
prevent this in-house bias, a stricter standard of Òbetter than bestÓ applies.
Independent financial advisers who are part of a company group cannot
recommend a product of the marketing group with which they are associated
5 ÒPrinciples and core rules for the conduct of investment businessÓ SIB 1991.
6 The independent channel is comprised of a number of different types of intermediary ranging from
Independent Financial Advisers currently regulated by the PIA to independent intermediaries that are
regulated by their professional bodies, i.e. accountants and solicitors. As IFAs are responsible for the
majority of business through the independent channel we use IFA and independent intermediary
interchangeably and note whether different types of independent adviser are differentially affected.
8
London Economics
July 2000
unless the product is demonstrably better than any other available product of
its type. In practice, this means that the IFA arms of providers do not sell any
in-house products.
3.2.2 Tied advice

Advisers may be tied to a single firm or to two or more firms that form a so-
called marketing group. Marketing groups are usually firms with ownership
links, but this is not necessarily the case. Tied advisers can be either:
• appointed representatives of a product provider; or
• company representatives employed by the providers.
The representatives of a product provider (or its marketing group associate)

must not advise customers to buy packaged investment products which
originate outside the marketing group. With tied advice, the training and
competence responsibility lies with the company tied to the adviser.
The definition of marketing group7 is relevant to polarisation because tied

advice must point the consumer to the most suitable product available in the
marketing group. For example, if company A enters a marketing group
which encompasses both company B and company C, the representatives of
company A would need to be able to advise on the most suitable product
from the whole marketing group. Reciprocity also applies, so that company B
and C representatives also need to advise on all products in the same
marketing group.
3.2.3 Exceptions to polarisation

There are a number of important exceptions to the polarisation rules.
The first is that they only apply to certain products. They do not apply, for
example, to pure protection products, or to mortgages. These products are
however often sold by agents who are also selling packaged investment
products for which their advice role is polarised. The polarisation regime
therefore allows IFAs to be tied to one or more companies for the purposes of
advice on products other than packaged investments, and allows tied agents
to advise on the non-packaged investment products of companies outside
their marketing group. Polarisation thus does not achieve a completely clear-
cut division between independent and tied distribution, but incorporates a
product borderline.
7 A marketing group corresponds to a group of companies that are (formally or informally) allied
together for the purposes of marketing the packaged products of the group.
9
London Economics
July 2000
Second, they do not apply in Òexecution-onlyÓ situations where only

information, and not advice, is being provided8. So for example, a fund
supermarket set up by a fund manager simply providing an execution only
dealing service could provide that managerÕs products among a range of
others. However, it would be subject to the Òbetter than bestÓ rule if, as an
IFA, it wanted to provide advice on these products9.
Third, they do not apply to stockbrokers and others undertaking

discretionary portfolio management. Discretionary managers undertake a
monitoring responsibility on behalf of the customer10.
3.2.4 Room for manoeuvre in the polarisation regime

Companies meanwhile can adopt a number of strategies in association with
other providers that come close to tying, but stay within the polarisation
rules.
• Outsourcing. Activities involved in the manufacture of the product

can be outsourced. So for example, a provider could outsource the
administration of a number of its products to another provider, or it
could outsource its fund management activities (or both).
• Multi-manager products. In these products, the customer can choose

among a number of external fund managers, whose brand is clearly
visible. These product offerings have been quite successful in recent
years.
• Reinsurance. Providers can manufacture their own product but re-

insure it with another provider: This is the method used by
companies to introduce a new with-profits product into their
portfolio. In terms of the consumer the product is wholly the
responsibility of the manufacturer. However, the product risk (in
this case intergenerational risk sharing) is borne by an outside
company.
All of these degrees of freedom are used by some tied providers, and relax
the rigour of the polarisation regime for them.
8 Execution-only refers to products sold without product specific advice. This needs to be distinguished
from direct offer promotions which are considered to contain product-specific advice and hence are
subject to the polarisation rules.
9 There are modified rules for introducers who gather data on individualsÕ circumstances for the
purposes of referring them to tied agents or independent advisers but do not provide investment
advise.
10 Independent Intermediaries can also provide Òbroker fundsÓ, life policies or contracts in an authorised
unit trust scheme whose value is determined by investment decisions made on a discretionary basis by
the adviser.
10
London Economics
July 2000
There is some confusion over the use of the term Òwhite-labellingÓ. This is
often used to describe providers using outsourcing or re-insurance
arrangements. In this case these arrangement do not change the providerÕs
responsibility to the consumer for the product and these are permitted under
the polarisation rules. However, it is also sometimes used to describe where
a provider re-badges another providers product and does not take
responsibility for the product. This is not currently allowed under the
polarisation rules. Because of this confusion we do not use this term
throughout the rest of the report.
Similarly, there is a degree of freedom within the polarisation rules for IFAs,
whereby they are able to establish panels of preferred providers to whom
they route a large proportion (if not all) of their business. This has allowed
the operation within the polarisation rules of IFAs who preferentially route
business to as few as 4 providers. The key difference between these
arrangements and tying with providers is that the IFA selects providers for its
panel periodically on the basis of the quality of their products and their
suitability for the IFAsÕ clients. The IFA may negotiate preferential terms
with the providers on its panel (e.g. for product enhancements) but remains
free to change its selection of providers at will. If more than 20% of an IFAÕs
sales are the products of a single provider it must inform the FSA in its
regulatory reporting.
3.2.5 Complementary regulation

Besides the polarisation rules, a number of other features of the regulatory
system also affect the provision of packaged investment advice. These
include:
p Suitability: the recommended product type must be appropriate for the

clientsÕ needs (age, job characteristics, preference for risk, existing
financial product holdings, etc).
p Product disclosure: the customer should be provided with key features

document, terms of business letter, etc.
p Commission disclosure: the remuneration earned by the adviser for

selling the product must be disclosed.
p Disclosure of tied adviserÕs remuneration: this represents the equivalent

cost of advice for consumers purchasing through tied channels.
p Training and competence requirements for financial advisers.
The complex and interlocking nature of the rules makes it quite difficult to
disentangle analytically the effect of polarisation alone.
11
London Economics
July 2000
3.3 The DGFTÕs 1999 report
The DGFTÕs report on the rules of the polarisation of investment advice,

published in August 1999, is a review of the operation of the polarisation
rules undertaken to:
Òestablish whether the rules have obtained the objectives for which they were
designed, whether they have enhanced consumer welfare, and whether they have had
any significant effect on competition in the market.Ó11
The DGFT looked at additional factors which influenced the operation of

polarisation, conducted interviews with product providers, financial advisers,
representative bodies and the ConsumersÕ Association, received written
responses, and looked at consumer research.
The report concluded that competition is restricted or distorted to a

significant extent by polarisation, through the prevention of some innovation
in retail markets, but that these effects are only significant in the market for
collective investment products. In the case of other investment products the
negative competition effects were thought to be less significant, leading to
these effects being outweighed by the benefits of polarisation.
The report recommended that:
p The polarisation rules be retained, but only for advice on investment

products linked to life insurance, including personal pensions. For these
products IFAs would remain regulated as today, with the
recommendation that a separation between the selection of products for
product panels and the negotiation of commission be required. It was
also proposed that Ògap-fillingÓ by lead providers be allowed, with the
provision that they assume responsibility for the product.
p For collective investment products polarisation be relaxed. The regime

would involve disclosure by all investment advisers both of ties and of
commercial arrangements for the newly non-polarised products.
p Greater transparency be introduced into the commission process, with

promotion of the potential to negotiate for a commission rebate, and the
neutralisation of taxation between advisers offering commission and
those offering fee-based advice.
11 The Rules on the Polarisation of Investment Advice, A report by the Director General of Fair Trading,
OFT, August 1999.
12
London Economics
July 2000
3.4 Regulatory approaches in other countries
While circumstances of countries differ, it is useful to put the UKÕs approach

to regulation of advice in an international context. In particular, we have
looked at four countries: the USA, Australia, Germany and the Netherlands.
The case studies were chosen for the differences in the regulatory regimes in
terms of the importance of:
• product regulation;
• different forms of disclosure; and
• alternative restrictions on the status of advisers.
The USA has a complex system of financial advice regulation, with

responsibilities split between State and Federal regulators. There are strict
disclosure rules about business volume which apply to advisers, but no
restrictions on multi-ties. The USA does not place any specific restrictions on
the use of the word ÒindependentÓ with regard to financial services. Instead,
advisers use competing sets of qualifications and designations.
Agents employed by companies such as Merrill Lynch and American Express

offer customers both in-house and externally manufactured products and in
many cases the share of external products in total business is substantial.
Many major firms undertake to be commission neutral between in-house and
external products. There has been a review of the impact of commission on
the sales process, but it is currently only Òbest practiceÓ rather than necessary
to be commission neutral.
While Australia does not have polarisation, it has a number of striking

similarities to the UK. In particular, it suffered from its own mis-selling
problems in the late 1980s, with accusations of over-charging, bad selling
practices, and misunderstanding of the meaning of ÒindependenceÓ.
Australia introduced disclosure of fees, charges and the financial interest of
the agents in the 1990s, and has strict restrictions on the use of the term
ÒindependentÓ in the financial advice arena.
Germany by contrast has very lenient regulation of the provision of financial

advice. For example, no qualifications are required for an individual to
operate as a Òfinancial adviserÓ, and there are no restrictions on which
products can be distributed by advisers, as long as the products themselves
are permitted by the relevant authorities. However, Germany imposes
considerable product regulation. The result is that very little independent
advice is provided.
The Netherlands does not place restrictions on ties, but instead focuses on
disclosure of the nature of the ties (shareholding, dealing, IT). The industry
says that independence does not lie in considering all possible products, but
13
London Economics
July 2000
in the freedom to select the main provider and change this choice over time.
In the Netherlands, historically it is fee-based advice that has been viewed as
detrimental, rather than commission. A possible justification would be that
commission gives advisers incentives both to sell products or to offer
appropriate products.
It is apparent from these international comparisons that the UKÕs polarisation

regime is unique. Other countries tend to protect consumers through use of
disclosure or product regulation rather than status restriction. While
different regimes will suit the different market circumstances of each country,
and the nature of the past regulatory regimes in each country partly
determines what directions of change are most feasible, the international
comparisons do allow us to observe how other markets have performed
under different rules. In our assessment of alternative regimes for the UK, we
have attempted to use these outcomes (whilst taking into consideration the
differences in the market structure) as supporting evidence.
14
London Economics
July 2000
Section 4 The market for polarised packaged products
4 The market for polarised packaged

products
The market for packaged investment products has grown substantially since
the polarisation regime was introduced in 1987, though year on year growth
has varied considerably. A combination of factors12 led to a fall in sales in the
mid-1990s, but the market has again grown strongly in recent years.
Meanwhile, the mix of the products sold has changed substantially:
• there has been a substantial trend towards investing in single

premium products; and
• the growth in retail unit trust products has far exceeded that of life
and pensions.13
Meanwhile the Government has introduced a number of new tax-privileged

savings vehicles.
4 . 1 Market shares of different distribution

channels
4.1.1 Market shares by new business premium

The market shares by value of different distribution channels for the life and
pensions and unit trust markets (including PEP and ISA sales) are shown in
Figure 1 and Figure 2 respectively.
Figure 1 shows that within the last few years, independent intermediaries
have overtaken tied agents to become the largest single channel in terms of
new business premiums for life and pensions business. Direct14 sales of life
and pensions products have remained very small, comprising less than 5% of
the market throughout the period.
12 These factors include: the introduction of commission disclosure, publicity regarding pension mis-
selling and the lack of confidence in the economy.
13 Retail investment in unit trust and OEICs have grown from £4.2 billion to £28.5 billion over the period
1992 to 1999, representing an annual growth rate of 31.2%. This is considerably faster than the annual
growth rate for life and pensions business. Over the period 1992 to 1999, the life and pensions business
grew from £4.5Êbillion to £5.9 billion, representing an annual growth rate of 4%.
14 It is important to differentiate between direct sales and direct sales forces. Direct sales refer to sales
made through mail shots, newspaper advertising, telesales and use of the internet.
15
London Economics
July 2000
Figure 1: Distribution of life and pensions new business (APE) 15,

£ millions, 1989-1999
100% Direct
80%
% of life and pensions business
Tied agents
60%
40%
20% Independent
intermediaries
0%
1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999
Source : ABI Insurance Statistics Yearbook
IFAs have a significantly higher share in the sales of rapidly-growing single

premium policies than they do for regular premium policies. There has been
a similar trend away from tied agents and towards IFAs within both premium
types.
Figure 2 shows that IFAs have not increased their share of the unit trust
market. The sales they make are predominantly unit trusts in PEP or ISA
ÒwrappersÓ, and they have a very low share of the investment trust saving
scheme market.16
15 The value of new business is calculated using the standard Annualised Premium Equivalent formula,
which is equal to the value of annual premiums plus 10% of single premiums.
16 As investment trust savings schemes represent only a very small share (approximately 1%) of the flows
into the retail investment funds we do not focus on this part of the market.
16
London Economics
July 2000
Figure 2: Distribution of unit trusts and ISAs gross sales (APE), £ millions,
1994-1999
100%
Private client
80% Direct
% of unit trust and ISAs
60% Tied agents
40%
Independent
20% Intermediary
0%
1994 1995 1996 1997 1998 1999
Source : AUTIF
The introduction of ISAs to replace PEPs in April 1999 has not substantially
altered this picture. Although independent intermediaries initially
concentrated on selling PEPs before the end of the 1998-1999 tax year in order
to take advantage of the tax benefits that would cease to exist post 6 th April
1999, they have retained their market share in the ISA market.
4.1.2 Market shares by number of customer purchases

If instead of looking at new business premiums we consider the number of
people buying through different channels, the market for packaged products
looks very different. Using data from the NOPÕs Financial Resources Survey,
Figure 3 and Figure 4 shows the proportion of people buying personal
pensions and life policies respectively through different channels from 1993
to 1999. This shows only a relatively small increase in the share of the IFA
channel.
17
London Economics
July 2000
Figure 3: Breakdown of the sale of personal pensions

by main distribution channels, 1993-1999
100%
Others
Tied agent
80%
Insurance
60% Company
40%
Bank or building
society
20%
Independent
intermediaries
0%
1993 1994 1995 1996 1997 1998 1999
Source: NOP Financial Resources Survey
Figure 4: Breakdown of the sale of life products by main distribution

channels, 1993-1999
100%
Others
Tied agents
80%
Distribution of sales
Insurance company
60%
40%
Banks or building
society
20%
Independent
intermediaries
0%
1993 1994 1995 1996 1997 1998 1999
Source: NOP Financial Resources Survey
18
London Economics
July 2000
According to this evidence, the IFA channel has a considerably smaller share
of the market when measured in terms of the number of consumer purchases
rather than in terms of new business premiums. This means that the average
premium size is considerably larger in the IFA channel than in the tied
channel and that the disparity has increased considerably over time.
4.1.3 Types of consumers

Although there are systematic differences in the types of consumer using
different channels, the differences are not as great as often believed. For
example, there is only a small difference in the average ages of those using
the various channels. The most significant differences are that customers
using IFAs or making direct purchases:
• are more likely to come from higher socio-economic groups. Thus

35% of IFA customers come from socio-economic groups A and B,
compared to only 15% for insurance companiesÕ customers17; and
correspondingly
• are more likely to be in higher income categories. Thus 15% of those

using IFAs earn in excess of £50,000 pa, compared to only 8% of those
using banks or insurance companies.18
This confirms the impression that IFAs have been successful in raising the
average level of premium by appealing to those with more to invest. That
success of the IFA channel has not, however, drawn a large number of less
well-off people into this channel as a substitute for tied channels.
4.2 Price trends
A useful proxy for the competitiveness of the market is to examine trends in

the price at which products are provided. An increasing price trend can
indicate increasing competition problems, while conversely a decreasing
price trend suggests that competition is becoming more effective. We use
Reduction in Yield (RIY) as a measure of the level of product charges. RIY is
the difference between the actual yield on the product and the yield the
customer would have enjoyed had there been no charges at all.
As reported in the 1999 PIA Disclosure Report, there has been a steady trend
to lower charges in life and pensions products. This evidence is reproduced
in
17 NOP Financial Resources Survey.
18 Continental Research.
19
London Economics
July 2000
Table 1: The changing reduction in yield (RIY) on life and pension

products, 1995-1999
Type of policy Average RIY Improvement

in RIY
1995 1999 1995-1999
10 year with-profit endowment 3.4% 3.0% 12%
Tax exempt 10 year savings plan 3.7% 3.0% 19%
25 year mortgage type endowment 1.7% 1.3% 24%
25 year personal pension 1.9% 1.6% 16%
Single premium bond held for 10 years 1.6% 1.3% 19%

Source: 1999 Disclosure Report, PIA
In addition to this downward trend in charges, there is evidence that the

structure of commissions is changing from up-front (or indemnity
commission) towards level loading (or trail commission). This is most
evident in the pensions market.
It is widely agreed that one of the largest sources of these effects is the
forthcoming (April 2001) introduction of the stakeholder pension, a personal
pension product which will have:
• a cap of 1% on its annual management charge; and
• no charges for starting and stopping contributions.
Whether an individual stays in the scheme or leaves early, the RIY will thus
remain at 1%. The contrast with the average level of early-leaver charges in
the personal pension market is very marked. In July 1999, individuals
withdrawing after 2 years from an average £200 per month regular premium
unit-linked personal pension faced an RIY of 13.6%. The average on the
market in 1994 was an RIY of 70%19, and as Figure 5 shows, the most of the
reduction since the early 1990s has come in the last year.
19 Money Management, Feb. 2000, p. 56, Table 2.
20
London Economics
July 2000
Figure 5: The changing reduction in yield for personal pensions, 1992-1999,

(25-year unit-linked personal pensions, £200 per month contribution)
120%
100%
1992
Reduction in yield
80% 1993
1994
1995
60%
1996
1997
40% 1998
1999
20%
0%
0 5 10 15 20 25 30
Years since policy taken out
Source: Money Management and London Economics calculations
4.3 The provision of advice
4.3.1 Independent Financial Advisers

Consistent with IFAÕs success in increasing average premium size, the large
increase in their volume of business has not been matched by a large increase
in the number of advisers employed in the sector. In 1999, there were around
25,000 registered IFAs, only slightly higher than immediately following the
introduction of polarisation. However, the number of IFAs today does
represent a recovery since the early 1990s, at which point the number had
fallen to between 20,000 and 21,000.20
Meanwhile, the structure of the IFA sector has changed in a number of

important respects. In particular, there was considerable growth in the
number of IFAs who were members of IFA networks during the mid 1990s.
Reflecting this trend the number of IFA firms has decreased dramatically
from approximately 8,500 to 4,000 since the early 1990s.
20 Consumer Panel Report 1998, PIA.
21
London Economics
July 2000
4.3.2 Tied agents

In contrast to the IFA sector, the number of people employed as tied agents
has fallen dramatically. Before polarisation it was believed that there were
over 200,000 tied agents. This has declined to approximately 60,000 today.
Looking at the total numbers employed in tied sales force is, however,
somewhat misleading. The large fall partly reflects the reduction in the
number of backroom staff registered. These staff are not directly involved in
advising customers day to day, and as registration requirements have
increased, it has made sense for companies to register those individuals only
where absolutely necessary.
The aggregate numbers also conceal the degree to which providers have
changed the method by which they distribute packaged investment products.
Initially, banks and building societies chose largely to remain independent.
Within the first four years after polarisation, however, most of the larger
players had tied to another provider or set up their own life office. In many
cases, an IFA arm was kept to service sales from pre-existing relationships or
to service high net worth consumers. A fair number of smaller building
societies who tied to a provider have since become introducers.
4.4 Cost of advice
In addition to the overall level of charges discussed above, the cost of advice
to consumers and how it is remunerated are also relevant to the effects of
polarisation.
4.4.1 Fee versus commission

The market for fee based advice is still relatively small, and the great majority
of IFA firms receive less than 10% of their income through fees.21 This varies
significantly between different types of independent intermediary, with
accountants and solicitors much more likely to be on a fee for service basis.
Although there has been a small increase in the volume of business through
fee-based channels there is no evidence to suggest that this market is likely to
grow dramatically in the short run.
4.4.2 Level of commission

PIA Disclosure Reports give five years of data on commission and cost of
advice for life and pensions products.22 Among the products for which there
21 ÒUK Independent Financial Advisers 2000Ó, Datamonitor analysis of Matrix Data statistics.
22 The PIA Disclosure Reports only included collective investment schemes in 1997.
22
London Economics
July 2000
are five years of data, commission rates are consistently lower for IFAs than
the equivalent sales cost to tied agents. There is considerable variation
between company representatives and appointed representatives in terms of
level and the trend.
Figure 6: Average five year cost of advice via different distribution

channels23 (£)
10-year endowment/savings plan

With profits Unit linked
1995 1996 1997 1998 1999 1995 1996 1997 1998 1999
600 600
Five year total cost of advice (£s)

500 500
400 400
300 300
200 200
100 100
- -
IFA AR CR IFA AR CR
25-year personal pension plan

With profits Unit linked
1995 1996 1997 1998 1999 1995 1996 1997 1998 1999
1,000 1,000
800 800
600 600
400 400
200 200
- -
IFA AR CR IFA AR CR
Note: AR = Appointed representative CR = Company representative

Source: PIA Disclosure Reports
On the whole, data on the levels of commission show a small reduction.

There is also evidence of an increasing trend to commission rebating amongst
IFAs as shown in the ÒreducedÓ terms column of Table 2 below. Under
commission rebating, IFAs pass through some part of their own commissions
to customers, often in the form of increased allocations of funds to the
underlying saving product.
23 It is important to note that these refer to average commission rates. Large IFAs and IFA networks may
negotiate higher levels of commissions.
23
London Economics
July 2000
Table 2: The extent of commission rebating

(percentage of new regular premium personal pension business written)
Commission terms
Full Reduced Zero
1999 68% 17% 15%
1998 68% 17% 15%
1997 76% 15% 9%

Source: 1999 PIA Disclosure Report.
To sum up, the quantitative evidence accords with the view of a number of
industry participants to suggest that competition has been working relatively
well in the advice market in recent years, with consequent benefits to
consumers.
There is evidence however that tied and IFA channels serve different types of
customers. As we have seen, they serve somewhat different socio-economic
groups, with IFAs having greater penetration at the top end of the market.
The growing market share of IFAs appears not to have been associated with
increased penetration in the middle and lower ends of the market.
24
London Economics
July 2000
Section 5 The economic effects of polarisation
5 The economic effects of polarisation
Polarisation has the potential to create two sorts of economic effect. It is

designed to help reduce consumer detriment in the market for packaged
investment products and the associated investment advice. It can also, like
other regulatory restrictions, create competitive distortions in the markets it
affects.
The market for packaged investment products (particularly those for the
more complex and long-term products) is often associated with high levels of
consumer detriment (defined below). Polarisation is only one of a set of
possible rules to try to minimise detriment. It clarifies the status of advisers
by preventing tied agents representing themselves as independent. In
principle, this might or might not be effective.
p If polarisation has no effect in preventing consumer detriment, then

relaxing it will have no adverse effect, and its economic pros and cons
depend solely on its competition effects.
p If it has some effect in preventing consumer detriment, then relaxing it

could have adverse effects, which need to be considered alongside its
competition effects.
5.1 Consumer detriment
Consumer detriment refers to a loss of consumer welfare, relative to what

would happen in a fully competitive market, that occurs when consumers:
• fail to purchase a product that they would if they possessed a full

information set;
• purchase a suitable product, but pay more than the cheapest price
available;
• purchase a product, but not the one which is most suitable for their
needs; or
• purchase a product believing it is of higher quality than it actually is.
These things happen to a certain extent in all markets, but become a matter of
policy concern when they are very common, or very large in scale. This is
only likely to occur in markets for products about which consumers are
relatively poorly informed. It is particularly likely with so called ÒcredenceÓ
goods, complex long-term products whose quality is not known at the
moment of purchase but can only be assessed a long time after the purchase
(if at all).
25
London Economics
July 2000
Packaged investment products are likely to be credence goods, as returns

from the product are uncertain and the implications of the charging structure
are not observable when they are purchased. Financial advice seeks to bridge
the information ÒgapÓ between product providers and consumers.
However, the provision of financial advice itself suffers from some of the
same informational problems. In particular:
• it is difficult for customers to assess the relative quality of advisers by

comparing them before purchasing; and
• the quality of the advice given often becomes apparent after a long
period (if at all).
It is therefore widely believed that complex, long term packaged investment

products and the advice on them have the potential for high levels of
consumer detriment.
One source of evidence on this point is the prominence of information

problems among the answers people cite when asked why they have not
purchased investment products.24 The next two graphs show the reasons
given by consumers who seriously considered buying a product but decided
not to (Figure 7) and those who had not considered buying a product (Figure
8).25
24 It is important to note that the FSA Better Informed Consumers report only surveys people who say
they are financial decision makers. These are not general population surveys and only cover
approximately half the population.
25 In both these questions consumers could provide multiple answers.
26
London Economics
July 2000
Figure 7: Reasons given for not buying a product (percentage of those who
seriously considered buying but did not buy)
30%
25%
20%
15%
10%
5%
0%
Lack of money Information related Product was a waste of Still considering Not the right time to buy
reasons money/not a good deal
Source: FSA Better informed consumers report, April 2000
Figure 8: Barriers to buying financial products for inactive consumers

(percentage definitely agreeing or tending to agree)
Too young - will make plans when

20%
older
Don't have enough time to make
32%
financial plans
I'll do it one day 37%
Already have enough savings and

39%
investments
Nervous about visiting a financial
43%
adviser
Future a long way off 44%
Can't be bothered 46%
Too old to take out new products 52%
Don't know where to start to make a

57%
decision about where to put the money
Financial matters too complicated 64%
Don't know enough about financial

70%
matters to choose suitable product
Not enough spare money 77%
More immediate need to spend money 77%
0% 10% 20% 30% 40% 50% 60% 70% 80% 90%
Source: FSA Better informed consumers report, April 2000
27
London Economics
July 2000
The most commonly cited reason for not purchasing was lack of money.
However, 17% of those who considered making a purchase cite information
problems as the reason for not doing so. At least 50% of inactive consumers
also give information problems as the main reason for not purchasing (citing
reasons such as not knowing enough, not knowing where to start, or finding
the products to be too complicated).
It therefore seems likely that information problems are causing considerable

consumer detriment problems in these markets. The advice market which
exists to bridge this gap is not preventing a proportion of consumers from
feeling unable to purchase financial service products for informational
reasons.
We now consider the other types of consumer detriment and their likely
scale.
5.1.1 The scale of detriment

Useful evidence is whether consumers are receiving the type of advice they
think they need. If we look at those who have purchased a packaged
investment product recently, it is clear that not all consumers are seeking
advice. As our survey shows (Figure 9), those not seeking advice make up
almost 30% of those using an IFA, and larger proportions in the other
distribution channels.
28
London Economics
July 2000
Figure 9: Demand for advice by distribution channel
100%
90%
80%
70%
60% No - just arrange the

transaction
50%
Yes - did seek advice
40%
30%
20%
10%
0%
IFA Bank Insurance Direct
company
Source: Continental Research. Base: All respondents26
Fewer consumers seek advice on simpler products, and our survey results
show that only around half of consumers are seeking advice when they buy
unit trusts, PEPs and ISAs. Even for products that are thought of as more
complex, such as personal pensions, over 1/3 of people only wanted the
transaction arranged. Nevertheless, the majority of consumers are looking for
advice27.
A majority of consumers also appear to understand the status of their adviser,

that is whether they are tied to one company and can only advise on its
products, or whether they are an IFA and can advise on any product on the
market (Figure 10).
26 Direct refers to dealing with the product provider directly through either post, phone or internet.
27 This represents an upper bound on the proportion who are seeking ÒspecificÓ investment advice. On
further questioning many of these consumers were seeking general advice on the details of a particular
product (this is especially true in the direct channel).
29
London Economics
July 2000
Figure 10: Extent of understanding of polarisation status by recent

purchasers of financial products, 1998
Proportion saying that their adviser was
Tied Independent In between/don't know

90%
80%
70%
60%
Percentage
50%
40%
30%
20%
10%
0%
IFA clients Life office clients Bancassurer
Source: Consumer Panel Report, PIA, 1998.
However, it is also clear from Figure 10, that roughly 20% of consumers do
not understand what type of adviser they are seeing.28 Those who purchase
products through life offices or bancassurers, but nevertheless believe that
they are obtaining independent advice, are particularly likely to suffer from
consumer detriment. This is consistent with evidence from more recent
research by the FSA (illustrated in Figure 11 below). This shows that between
20% and 30% of consumers thought their adviser was tied to more than one
company. As polarisation prevents advisers tying this way, it appears that
these consumers did not understand the status of their adviser.
28 Qualitative survey evidence suggests that the proportion may be even higher than this.
30
London Economics
July 2000
Figure 11: Consumers perception of the independence of their financial

adviser (by investment product)
Life assurance 55% 18% 27%
Tied to a single company
Personal Able to advise on a few

52% 30% 18% companies
pension
Able to advise on any
company
Investments 49% 19% 32%
0% 20% 40% 60% 80% 100%
Source: FSA Better informed consumers report, April 2000, Base: All products bought or seriously considered
through an advisor
Finally, there is some evidence that confusion over status is highest in the
bancassurance part of the market. The next figure shows that 31% of
consumers believe that banks are the best place to get independent advice,
even though banks largely have tied distribution.
31
London Economics
July 2000
Figure 12: Asked if Òa bank is the best place to get independent financial
adviceÓ
70%
63%
60%
50%
40%
31%
30%
20%
10% 6%
0%
Agree Disagree Don't know
Source: Consumer Panel Report, PIA, 1999. Base: All financial decision makers
This is confirmed by the survey undertaken for this project, which found 39%
of those going to a bank were unaware when they first went to them that
banks could offer only their own products.
The evidence suggests a relatively high level of understanding. An important

caveat is that these surveys are of recent purchasers or financial decision
makers. It is possible that understanding is considerably lower among
consumers who have not been through the sales process.
Consumer inertia
Problems of consumer detriment will be mitigated to the extent that

consumers re-evaluate their choice of adviser and periodically compare
different types of advice.
Conversely, detriment is more likely to occur when consumers are ÒinertÓ,

and remain with whatever distribution channel they start using. These
consumers may be facing high switching costs, and profit-maximising firms
will inevitably tend to exploit this fact.29 It is helpful to differentiate between:
29 Little shopping around could also reflect that there is known to be very little price variation.
32
London Economics
July 2000
• active consumers, who choose the distribution channel themselves;

and
• passive consumers, who are approached by the representative or

provider that they eventually bought from.
The balance between these types is likely to have large implications for how
changes in the polarisation rules would change consumer behaviour.
There are some important indications that most customers are passive. The
PIA Consumer Panel Report defines active consumers of packaged
investment products as those who regard themselves as Òhands onÓ, and
finds that only 16% of people display this approach to financial planning.30
These people agreed with statements such as ÒI enjoy finding out about new
savings and investment schemesÓ and ÒI like to shop around for the best
deals on interest rates, mortgagesÓ. While these people would probably not
become trapped in a channel, a large proportion of the population could be,
and thus become vulnerable to consumer detriment.
Consumer research for this project suggests that a rather higher proportion of
customers are taking the initiative. It shows that around 2/3 of recent
purchasers made the initial contact that led to their most recent purchase,
while only 1/3 were approached by providers. As shown in Figure 13, the
percentage of people who make the initial contact for themselves is highest in
the IFA channel.
30 Consumer Panel Report, PIA, 1998.
33
London Economics
July 2000
Figure 13: Proportion of consumers making initial contact

(by distribution channel)
100%
90%
80%
Company/person
70% made the initial
approach
60%
50%
40%
You made initial
30% approach
20%
10%
0%
Direct
Bank
IFA
Insurance
company
Source: Continental Research. Base: All respondents
In all channels, the percentage of people who made the initial approach for
themselves is over 50%. Figure 14 examines how this breaks down by type of
product. The percentage of people making the initial approach for themselves
is higher for the simpler products.
34
London Economics
July 2000
Figure 14: Proportion of consumers making initial contact (by product)
100%
90%
80%
Company/person
70% made the initial
approach
60%
50% You made initial

approach
40%
30%
20%
10%
0%
Unit Trust
Investment
ISA
PEP
Personal
pension
Endowment
bond
policy
It is also useful to compare people purchasing through a new channel (who

might be freer to shop around) with those with an existing relationship with
an adviser (who might be ÒtrappedÓ). Evidence on this is shown in Figure 15.
35
London Economics
July 2000
Figure 15: Proportion of consumers making the initial contact the first time
they used a distribution channel versus the most recent time
90%
80%
70%
60%
50%
First time
40% Used channel before
30%
20%
10%
0%
company
Purchasing through an IFA, bank or directly increases the probability the

consumer will make an active choice for further purchases. But customers of
an insurance company are more likely to be contacted by the firm after the
first purchase.
In conclusion, a large percentage of people appear to actively contact their

adviser and this proportion is higher for those purchasing simpler products.
Though consumers are therefore actively recognising the need for financial
advice, this is not inconsistent with consumers then being sold a particular
type of product or that of a particular provider. It is therefore reasonable to
conclude that the level of consumer detriment from consumers inertia is
likely to be lower for ÒsimplerÓ products.
Shopping around
There is concern that consumers do not shop around and this also exposes
them to detriment, since they do not have a clear idea of the prices and
characteristics of alternative products. PIA surveys reveal that just over 35%
of people look for at least a second opinion on their investment product
choices, less than for general insurance.
36
London Economics
July 2000
Figure 16: Extent of shopping around for advice on savings and investment
products, 1996 - 1998
70
60
Percentage of investors
50
1996
40
1997
30
1998
20
10
0
Did not shop around Two advisers Three advisers Four or more Not sure
Extent of shopping around
Generally we would expect to see more shopping around in markets Ð like

packaged investment products Ð where the financial commitment is large.
However, it appears that more people (about 45%) shop around for general
insurance products than for saving and investment products (35-40%). The
perceived complexity of saving and investment products appear to make it
more difficult for consumers to compare product offerings and hence lower
the value of shopping around.
Our own survey data reveals the extent to which consumers who bought
through one channel had also considered other channels. Figure 17 shows the
results.
37
London Economics
July 2000
Figure 17: Did consumers using one channel consider other channels?
100%
90%
Insurance company
80%
% considering other channels
70% Dealing direct
60%
50% IFA
40%
Bank/building society
30%
20%
Did not
10%
0%
company
Source: Continental Research. Base: All respondents.
It is clear that a large majority (over 75%) only look within a particular
channel. As a result, our sample of those who considered alternative channels
is relatively small.31. Looking within channels it is clear there is also relatively
little search between advisers. Of those using an IFA, 75% only see one IFA.
31 Within this small sample it is interesting to note that bank customers mostly see IFAs as the closest
substitute and IFA customers see banks as the closest substitute.
38
London Economics
July 2000
Percentage of respondents saying the source most influenced Figure 18: The influence of advisers
All those using an adviser Those using a tied adviser Those using an independent financial adviser
80
70
60
their decision to buy
50
40
30
20
10
Other
The internet
Friends/relatives
newspapers/magazines
Product information
recommendation
Adviser's
Information in
This may not imply that the consumer does not undertake any search at all, as
information can be gathered from other sources even if an alternative channel
was not considered for the purchase decision itself.
Research published in the 1999 PIA Consumer Report (shown in Figure 18)
shows that 40% of consumers cite the adviserÕs recommendation as the most
important one in their decision to purchase, suggesting that Ð once they have
chosen an adviser Ð consumers are quite reliant on the advice he or she
provides. This is dramatically higher for those using an IFA.
To sum up, the high level of initial approach suggests a considerable number
of active consumers, but the subsequent low level of shopping around
suggests they may still face high levels of consumer detriment. The simpler
the product, the greater are both the level of initial contact and the level of
shopping around and thus the less likely are consumers to suffer from
consumer detriment.
5.1.2 Does polarisation affect detriment?

Changing the polarisation rules could affect detriment by making consumers
more confused about the type of adviser they are using. To some degree the
behaviour of consumers lessens this concern:
• more consumers than previously thought actively contact their

adviser;
39
London Economics
July 2000
• status disclosure has proved relatively successful, in that a majority

of consumers understand the status of their adviser and use the type
of adviser that is consistent with their perception of their needs; and
• the degree of reliance on the adviserÕs recommendation differs

between tied and independent channels reflecting the difference in
the type of advice provided.
However, the potential for detriment is increased by:
• the limited extent that consumers re-evaluate their choice of adviser;

and
• the lack of shopping around for advisers between channels and

within a particular channel.
The fact that confusion Ð and thus the potential for detriment Ð exists in the
present regime does not necessarily mean that changing polarisation would
create greater detriment. Unfortunately, it is not possible to find statistical
evidence on the degree to which consumers understood the status of their
adviser before the polarisation regime was introduced. Anecdotal evidence
from providers working in the market at this time, however, suggests
consumer confusion was considerably higher than it is now.
5.2 Competition effects
Many of the arguments in favour of the relaxation of the polarisation rules

centre on its detrimental effect on competition. For the purpose of
competition analysis, polarisation rules can be considered as vertical restraints.
More specifically, the obligation on tied agents to distribute the packaged
investment products of a single product provider (or more accurately a
marketing group) constitutes what are known as exclusive dealing (a particular
kind of vertical restraint).
The competition problems resulting from exclusive dealing could materialise

in a number of ways:
1. A provider of packaged investment products may not be able to distribute its

products because it cannot sell through tied distribution channels.
Evidence from a variety of types of provider suggests that those without a

full range of products are unlikely to find any existing distributor willing to
tie to them. This is illustrated by:
40
London Economics
July 2000
• Evidence from friendly societies (who produce a product only they

are able to manufacture) that they are not able to find distributors
willing to tie to them.32
• Evidence from small providers, such as unit trust companies, that

they cannot get distribution through the tied sector.
Of course, it is also possible for a product provider to use its own tied
salesforce. However, there are strong economic reasons why it is unlikely to
be profitable to do so without a full product range. In particular:
• Evidence from those in the tied channel is that it is only efficient to

distribute through this channel with a product range that meets most
customer needs. The salesforce model conventionally builds up a
customer base and sales are made by regularly contacting the
customers until a financial need arises. The costs of maintaining such
a relationship with the client is best shared across a number of
products.
• The bancassurance model relies to a greater extent on consumers

approaching their existing financial services provider. As the
network represents a significant fixed cost, it is efficient to sell as
many types of product as possible. Only those with an existing
distribution network could hope to replicate this model.
2 . A distributor may not be able to distribute the products of a provider, even

though they are better than those of the provider it is tied to.
There is evidence from tied providers that they would be able to provide a
better portfolio of products if they could distribute products from a number
of providers in different product categories. For example, evidence from
providers suggests this results in:
• uncompetitive products being provided (for example, we were given

evidence that polarisation had resulted in poor value annuities being
introduced into the portfolio of a provider);
• suitable products but not necessarily the best product; or
• no product at all.33
32 Friendly societies have been successful in developing relationships with introducers. However, it is
their experience that if the relationship is successful the introducer will eventually look to tie to a
provider who offers a wider range of products.
33 If there is not a suitable product available tied agents should refer a customer to an independent
intermediary. However, providers believe this sometimes results in the customer not purchasing and
a financial need not being met.
41
London Economics
July 2000
In addition, if a distributor ties to a provider the tied adviser must give advice
over the entire range of products provided by the marketing group. If the
distributor focuses only on some parts of the market, e.g. small building
societies focusing on a limited set of mortgage related products, the cost of
providing advice might outweigh the resulting sales making it uneconomic to
provide any products at all. We were presented with evidence that this had
resulted in a significant reduction in the volume of sales.
3. A distributor is not able to manufacture its own product in a product category

and distribute products of other manufacturers in the same category.
Even where the manufacturer produces its own product it may wish to sell
that of other providers as well. This might reflect differences in efficiency
between different manufacturers resulting from specialisation or different
manufacturers targeting different customer types. Given that there are search
costs for consumers it is efficient for retailers to sell the products of multiple
manufacturers. This model is commonly adopted in other countries.
If these propositions are supported this could lead to a number of

competition problems:
• lack of competition and innovation as providers who are the most

efficient or the most innovative cannot enter the packaged investment
product market; and
• less competitive terms as those in the market find it easier to collude

with each other.
On the face of it, the competition problems associated with these propositions
are substantially weakened if:
p The tied channel competes with a competitive independent intermediary

sector (i.e. the tied and independent channels supplying the same
market).
p There is intense competition within the tied sector (i.e. the tied market is
competitive such that innovation and improvement in product terms
occur).
We discuss these possibilities in turn.
5.2.1 Independent intermediaries

If the independent intermediary market serves the same customers as the tied
market, competition from the independent market will discipline the tied
channel. The evidence on this is inconclusive but suggests it is unlikely that
the IFA and tied channels serve exactly the same market:
• The level of prices as represented by the average RIY in the two

markets are roughly the same. However, this comparison does not
42
London Economics
July 2000
take account of the level of commission rebating. This would lower

the average price of the IFA channel (see Section 4.4.2).
• Although there is considerable overlap in the types of consumers,

there is little evidence that consumers have switched channels (see
section 4.1.2) or that they compare channels (see section 5.1.1).
• Although the level of prices might be similar it is often argued that

the quality of advice is better in the independent channel as
evidenced by differences in persistency rates.34
While we cannot say with certainty whether the IFA and tied markets are in
the same ÒeconomicÓ market, the weight of evidence suggests that
competition between them is not perfect.
Because of the uncertainty, we investigate both alternatives. Firstly, what if

the independent sector does compete to serve the same customers. In this
case all providers are serving the same ÔeconomicÕ market.
Table 3: Market concentration (1993-1998)
1993 1994 1995 1996 1997 1998
A) Life Premiums
Top 5 25% 24% 23% 24% 23% 27%
Top 10 38% 38% 38% 37% 39% 43%
Herfindahl 241 238 236 267 258 290
B) Pensions Premiums
Top 5 31% 36% 37% 39% 36% 40%
Top 10 43% 51% 52% 56% 53% 58%
Herfindahl 302 388 434 456 421 440
Source: FSA Returns
The market shares of leading providers as a percentage of total industry

premium income give measures of concentration. These are summarised in
Table 3 for life and pension products over the period 1993 to 1998. This
reports the proportion of new business relating to the top 5 and top 10
providers, together with the Herfindahl index35 for that particular market
34 Evidence from the PIA Persistency surveys. There is concern that the differences in persistency only
represents the differences in the types of consumer. However, there is insufficient evidence to prove
this proposition.
35 The Herfindahl index provides a relative measure of market concentration, and is calculated as the
sum of each individual companyÕs market share squared It therefore has the potential to take on a
value between 0 and 10,000. The Herfindahl index will take on a value of approximately 0 in the case
of a industry with an infinite number of companies, and 10,000 in the case of an industry with just one
43
London Economics
July 2000
segment. These statistics are reported for APE, i.e. 10% of single premium
business and 100% of regular premium business.
Mainly through a process of consolidation, the provision of both life and

pensions has become more concentrated, with the levels of market
concentration in 1998 being noticeably higher than in 1993. The concentration
in the provision of pension products (where the top 10 providers did 58% of
new business in 1998) was higher than that in life products (where the top 10
did 43%).
However, although increased concentration is a source of potential concern,

the current level of concentration is not in itself sufficient to worry
competition authorities. For example, the US anti-trust authorities regard
Herfindahl indices of over 1000 as a matter of possible concern, and none of
these markets shows a Herfindahl of above 500.
Table 4: Concentration in gross unit trust sales
Top 5 33%
Top 10 47%
Herfindahl 334
Source: UK Fund Industry Review and Directory, 1999
Evidence on the provision of unit trusts (Table 4) suggests a similar low level
of concentration.
This evidence suggests that the product market is not very concentrated,
whether it is competitive depends on whether a new provider could
distribute through the large number of IFAs (currently providing over 50% of
the market in terms of premiums). If they will consider a new product
provider, it is reasonable to conclude that there are low barriers to entry for
providers.
Is access to the IFA market competitive?
Concern has been expressed that new providers cannot get IFAs to distribute
their products. In particular, this has focused on the expanding IFA networks
and the increased importance of product panels. Networks showed steady
growth in the mid 1990s, although this has slowed slightly in the past couple
of years. The use of panels has also increased and it has been estimated that
panels are used in 80% of cases.
company.
44
London Economics
July 2000
IFA networks are now estimated to have 40% of the IFA market and 22% of
the total financial advice market.36 The largest network, Misys, now has over
4,000 IFAs giving it approximately 17% of the IFA market and 9.5% of the
total financial advice market.37 This position gives the larger networks a
position unrivalled by any tied company.
There are signs that concentration is allowing IFA distributors to exert market
power, but little evidence that the gains are going to be passed onto
consumers. Recently some networks have been asking for commission levels
10% above the standard national and regional IFA rates.38 Although other
networks appear to be less demanding, their own testimony is that they
would not accept lower commission levels than those of other networks.
Meanwhile, the OFT recently gave competition clearance for Misys to acquire
ie group, increasing its share of the IFA market.39
The use of panels, however, is largely supported by IFA providers. They

suggest that panels have largely enhanced competition through:
• greater consistency throughout the country;
• more transparent criteria on which to compete;
• centralisation of the decision making process making information

provision easier; and
• regular re-assessment of whether a product should be included.
The size of the panel and the speed with which it is updated are clearly the
key to whether a panel enhances or limits competition. Although, the
appropriate size of the panel will vary depending on the type of IFA, there is
some concern that very restrictive panels appear to be used.
Are there barriers to setting up as an independent intermediary?
Even if concentration in the IFA channel is rising, this only represents a

problem if there are barriers to the entry of new IFAs. The number of IFAs
has risen over the last 3 years (see Section 4.3.1). However, it is argued that
barriers to entry have risen considerably because of:
• increased training and competence regulation;
36 Assuming IFAs working inside and outside of networks have the same average premium.
37 Assuming IFAs working for Misys and the rest of the IFAs have the same average premium.
38 Money Marketing, May 18th, 2000.
39 Money Marketing, June 15th, 2000.
45
London Economics
July 2000
• reduced numbers of tied agencies that provide the traditional

training ground for independent financial advisers;
• increased costs of professional indemnity (PI) insurance; and
• increased capital needs as the result of the move from indemnity

commission to level loading.
The last of these can lead to substantial problems in entering the IFA market,
making the provision of contacts offered by a tied network more attractive or
changing the focus of business into areas where a more traditional
commission structure has been maintained (e.g. protection products).
Small IFAs believe that this would not prevent entry, though it could lead
IFAs to change their focus or look for temporary assistance (such as initially
being a multi-tie if this was possible).
The evidence would therefore tend to suggest that, although there are some
causes of concern, the IFA market is broadly competitive. This should
alleviate the problems facing a new provider entering the market.
5.2.2 Distributing through the tied channel

If the independent channel does not compete for the same consumers, then
we need to consider the tied channel in isolation. If competition in the tied
channel is intense then this itself should lead to better terms and more
innovation.
Table 5: Concentration in life and pension products in the tied channel
1996 1997 1998 1999
Top5 68% 63% 54% 51%
Top10 85% 82% 72% 71%
Herfindahl 1418 1374 989 775
Source: Money Management and London Economics calculations
It is only possible to estimate quite roughly the market share in the tied
channel with the results shown in Table 5.40 It appears that concentration is
40 These figures were estimated from results of a Money Management survey. This does not correspond
to the whole market and therefore gives an upper bound estimate.
46
London Economics
July 2000
significantly higher in the tied channel. Although concentration has been

falling there is some concern over its level.41
As we have seen (section 5.1.1) consumers do not shop around within

channels. The difficulty in comparing products across providers means there
is relatively little competition for these customers.
The problems of product comparison therefore may stifle competition. It is

difficult for tied channels which improve their product offering to gain
market share, because few if any consumers observe this improvement.
There is some evidence that some providers use the freedom available within
the polarisation rules to offer better products. A number of leading providers
do for example take the opportunity to outsource administration and fund
management. A number of providers use re-insurance or outsourcing
arrangements to provide products they could not otherwise provide
themselves. However, there is also evidence that the costs of doing this
sometimes prevents better products from being introduced.42
Allowing tied providers to introduce the brands of existing manufacturers

would enable them to better signal an improvement in product terms. There
is some evidence that this would be profitable and that they would take up
the opportunity. This evidence comes from developments such as recent
acquisitions of manufacturers who have traditionally focused on the IFA
channel by providers, the introduction of fund supermarkets, and the way
proprietary and non-proprietary products are distributed alongside one
another in tied networks in the US.
The weight of evidence suggests polarisation does slow down the speed at
which innovation takes place and blunts competition within the tied channel.
Therefore, the gains from removing polarisation in terms of competition are
likely to be positive, but modest in the short run.
5.3 Scenarios to be investigated
Polarisation is likely to impact on both consumer detriment and the level of

competition. Polarisation could be relaxed in a number of different ways, and
it makes sense to examine the policy scenarios likely to produce the greatest
net benefits in the light of this evidence. The costs in terms of detriment will
come mainly in the IFA channel (where consumers might once again be sold
products by tied advisers masquerading as independent intermediaries as
41 The appropriate market may, however, be much more narrowly defined, for example it may be more
appropriate to look at market share in annuities than life and pensions in general. In this case
concentration could be considerably higher still.
42 The additional cost imposed on re-insuring a product was estimated by providers to be between 30%
to 50% of the cost of introducing the product.
47
London Economics
July 2000
they were before polarisation was introduced). The least costly policy
scenarios in detriment terms will thus be those which do most to maintain a
clear independent status; i.e.:
• scenarios where polarisation is changed so that multi-tied advisers

are allowed to develop, but either are not allowed to claim that they
are independent, or are restricted to being in the tied channel. This is
in contrast to a Òfree for allÓ in which advisers can describe
themselves in whatever terms they please; and
• scenarios where certain products are exempted from polarisation, but

the distinction between IFAs and tied agents is maintained, as long as
this can be achieved without itself confusing consumers about
advisersÕ status.
To the extent that there are competition benefits, they are likely to arise in the
tied channel. The most beneficial scenarios are likely to be those that do most
to enhance competition in that channel; i.e.:
• scenarios in which barriers to entry into tied distribution are lowered,

to enhance competition between tied providers, for example by
allowing gap-filling; and
• scenarios where tied channels are allowed to develop multi-provider

offerings, potentially enhancing competition within the offerings of
each tied provider. As competition is relatively ineffective for
products where there are large information asymmetries, this should
be focused on simpler products.
These considerations underlie the design of the scenarios presented in Section

7 and analysed in Section 8 of the report and the Appendix.
48
London Economics
July 2000
Section 6 Technology and regulation
6 Technology and regulation
6.1 Technological change
New technologies have the potential to alter the way consumers gather
information about products as well as manage their existing products, and
could thus have important effects. E-commerce and the Internet already
provide consumers with the ability to manage and move funds online. These
changes affect polarisation in two ways:
• they increase the availability of information on financial services

issues, which may cut information asymmetry; and
• they cut the cost of setting up distribution, which may lower barriers
to entry.
At the same time, Internet technologies have a number of important

limitations for marketing packaged investment products. First, electronic
signatures and the laws concerning electronic signatures are only now
coming into place. More important, while current Internet technologies are
excellent at providing information they have so far provided quite limited in
providing advice. Because packaged investment products are complex,
individuals who use the Internet often require further advice. Indeed,
research quoted from Fidelity shows that the number of people in the USA
who ask for advice after using the Internet to obtain financial information has
risen.43 As Internet use for financial services expands in the UK, this would
suggest that there will be an increase in the demand for advice here as well.
The limitations of the Internet as an advised channel are illustrated by the fact
that technology is only used to deliver about 4% of advice, while the most
widely-used mechanisms for obtaining advice remain face-to-face and by
telephone (with 44% of people obtaining information in this way and a much
higher percentage favouring this method). This is shown in Figure 19 below.
43 Collette Bowe, Head of Western Pooled Funds, Fleming Asset Management, talk at the Polarisation
What Next? conference, 29th March 2000.
49
London Economics
July 2000
Figure 19: Preferred means of obtaining information on financial products
7%
Financial services regulation
1%
Method of obtaining information
18%
Teletext/Internet/CD-ROM
4%
14%
Advertisement
10%
31% Preferred
Family and friends
17% Actual
40%
Article/programme
24%
46%
Leaflet/booklet
27%
73%
Face to face/telephone
44%
0% 10% 20% 30% 40% 50% 60% 70% 80%
Percent
Source: FSA Better informed consumers report, April 2000, Base: Actual source: product bought/considered,
Preferred source: All respondents
Even fewer consumers currently purchase investment products on-line, but

there is a large potential for growth in this market. Advantages such as lower
search costs and lower transaction costs for the consumer 44 are leading to
innovation in the retail financial services market, with a number of new
models of distribution.
The obstacles to online advice may be lower for digital TV and other
broadband technologies that are in the pipeline. These media provide enough
bandwidth to enable 2-way video communication between advisers and
potential customers, making online advice a much closer substitute for face-
to-face communication.
6.1.1 Fund supermarkets

The number of UK fund supermarkets is increasing rapidly. Providers such
as Egg are already providing execution-only services. They are quickly being
followed into the market by competitors using a number of separate
strategies, the key component of which is whether it is the provider of the
supermarket or the fund that captures the resulting client relationship.
44 Resulting from the lower distribution costs for the product provider.
50
London Economics
July 2000
As noted above, the degree to which fund supermarkets lower the demand
for advice has been questioned by research in the United States. This research
suggests that the Internet is used primarily as a source of information after
which consumers may then have a more informed position from which to
seek advice, or may be confused by the proliferation of information and seek
advice.
6.1.2 Online advice and purchasing

Online financial advice is a new phenomenon in the UK, but one can look at
the first entrants into the market to gain some idea how this market might
develop. The website of Sort.co. charges individuals on a fee basis depending
on the type of advice required (i.e. whether it is pure investment advice or
advice on a pension). It mails/emails customised reports to individuals,
based on information provided online. Although it provides a link to a
discount broker, Sort.co.uk does not itself facilitate purchases. Importantly,
in the light of peopleÕs stated willingness to pay for advice (Figure 26) it is
currently able to provide advice for under £100.
Other online services in the UK tend to be more informational than advisory.

Interactive Investor allows individuals to pose specific questions to IFAs or
find an IFA online. There are also a large number of financial services sites
which provide information rather than advice.
6.1.3 Business to business E-commerce

The UK has not witnessed the same degree of use of electronic trading as the
US, but here too things are changing quickly. Soon IFAs will have access to:
• Systems provided by independent providers such as The Exchange.
• Systems developed with the markets trade association: EMX is an

electronic messaging system transmitting information between unit
trust and OEIC providers and IFAs. It will allow IFAs to place orders
with product providers and providers to confirm deals. EMX will aim
to facilitate comparison on short-term performance and charges.
• The mainframe computers of large product providers. This allows

the IFA to undertake more of the administration for the provider and
be rewarded for this service.
These developments could lead to significant by lower search costs for IFAs.45
Whether these gains are taken in the form of IFA profits, lower prices for
consumer or IFAs undertaking more frequent searches and passing on benefit
to consumers, is as yet unclear.
45 For a more detailed discussion of these issues see Business to Business e-procurement, Small and
medium sized enterprises, The Group Trade Report, May 2000.
51
London Economics
July 2000
6.2 Regulatory change
A number of important initiatives are underway to standardise products, as

well as the information available to consumers. Their aim is to lower
consumer detriment and encourage competition. It is important that the
polarisation regime be consistent and complementary with them.
6.2.1 Simplified products
CAT-marked products46
The TreasuryÕs initiative on CAT standard products aims to address the

information requirements of consumers by the creation of benchmark
standards for Charges, easy Access and fair Terms. It is intended that the
CAT standards will reduce consumer detriment by:
• addressing the consumerÕs information needs;
• making the product easy to understand; and
• providing a guarantee of a minimum level of product terms.
The Treasury has thus far introduced CAT standards for Individual Savings
Accounts (ISAs) and mortgages. It is not the stated objective that the
products should be sold without advice, but it may not be possible to provide
within the price cap.
ISAs, introduced in April 1999, were the first CAT marked product. There
has been relatively little evidence to date to show that CAT marked ISAs have
fundamentally changed the market,47 since:
• relatively few firms chose to introduce a CAT marked ISA;
• there is still considerable variation in the terms offered for non-CAT

marked ISAs and these dominate the market; and
• ISAs are perceived as being excessively complex as demonstrated by

consumers purchasing multiple maxi ISAs. 48
46 In addition to CAT marks imposed by the government, the life insurance industry is also working to
develop standards of its own through an initiative called SALTR. This initiative will result in voluntary
standards focused on reducing the scope for consumer detriment by adopting high standards on
literature/product information, charging structures and consumer service. The initiative will be
administered by an independent accreditation board.
47 However, there is a review currently being undertaken by McKinsey that should throw further light on
this issue.
48 Recent evidence suggests that only 33% of people find even cash ISAs straightforward, with 37%
52
London Economics
July 2000
Stakeholder pensions
The stakeholder pension is more restrictive in that only products complying

with rules will be able to use the brand Òstakeholder pensionsÓ at all. Indeed,
the main characteristics of a stakeholder pension (a 1% cap on the annual
management charge, and a degree of transferability) have already had a
major impact on the personal pension market.
The remarkable impact of stakeholder pensions can be explained by ruling

RU64 issued by the PIA in the spring of 1999. This prevented advisers selling
pensions that could materially disadvantage consumers on the introduction
of stakeholder pensions. This change has been associated with the average 2
year RIY falling from 37% to 14%.49
6.2.2 Comparative information scheme

Comparative information tables or Òleague tablesÓ are intended to provide
information on financial products such that consumers can compare one
providerÕs product to another.
The comparative information scheme will replace the existing disclosure

reports. However, they differ fundamentally in that they are aimed at
providing the consumer directly with information in a standardised format.
If the tables do facilitate comparison, the effect should be to:
• encourage shopping around;
• add an external reference to the advice provided by advisers; and
• increase competition around the characteristics included within the

comparative information tables.
This last point suggests one of the possible drawbacks of the comparative
information schemes Ð that the characteristics included in the tables may
become a focal point of competition, possibly to the exclusion of other
important product characteristics.
6.2.3 Decision trees

As part of its consumer information programme, the FSA has recently
indicated that it may introduce decision trees to aid the sale of stakeholder
thinking them complicated and 28% being unfamiliar with them. In terms of complexity, cash ISAs
ranked worse than all other products except equity ISAs, unit trusts and gilts, placing them as more
complex than personal pensions, endowments and stocks and shares amongst other products. Better
informed consumers, Consumer Research 1, Financial Services Authority, 2000.
49 Money Management, February 2000.
53
London Economics
July 2000
pensions. Although this is still at the consultation stage, the main purpose of
decision trees would be to increase the level of understanding regarding
stakeholder pensions and to provide consumers with information about what
options are available to them.
It is important to recognise that decision trees would not provide specific

advice. Indeed, the FSAÕs consultation document states that:
Òthey are intended to help consumers make a choice, where they reasonably can,
without having to pay for advice, and to help identify instances in which advice or
further information is necessaryÓ.50
The impact of decision trees could therefore be to increase the likelihood of

consumers using the direct channel, and to reduce the need for advice.
However, the trees are likely to end with the suggestion that further advice
might be necessary. This could actually be to the benefit of the IFA channel,
particularly since (it is suggested in the current draft) that contact details of
IFA associations will be provided.
6.2.4 EU regulation
Polarisation could be affected by two directives being developed at the
European Commission, the Insurance Intermediaries Directive and the E-
commerce Directive.
The European Commission has drafted an outline proposal, which will form
the basis of the new directive on the insurance intermediaries. It is hoped
that the Council of Ministers will adopt it by the end of the year 2002.
The current draft proposal has the objective of facilitating the freedom to
provide services across member states. Member states will be allowed to
impose additional requirements but only for the intermediaries established in
their own territories. In the case of intermediaries from other member states,
who wish to establish themselves, the mutual recognition of professional
capacity as set out in the 1976 directive would still apply but in a much
simpler manner. The largest effect of this could be to reduce barriers to entry
into the UK intermediary market.
The EU E-commerce directive,51 currently under discussion, also has the

potential to change the regulatory regime substantially, by subjecting a
company to conduct of business rules of the country that represents its Òbase
of activityÓ rather than in the country where the customer is located. As a
result, foreign firms providing packaged financial services products from
50 The FSAÕs approach to the regulation of the conduct of stakeholder pension business, May 2000.
51 The current draft of the E-commerce directive does not apply to UCITs. There is a derogation for UCIT
providers that they must comply with host country conduct of business rules. This derogation may
disappear in future drafts.
54
London Economics
July 2000
abroad over the Internet would cease to be subject to UK conduct of business

and polarisation rules in the future.
6.3 Conclusions
Technological change has a number of important implications for

polarisation. In particular, the growth of Internet e-commerce has cut the cost
of information relative to advice. However, the limited evidence so far
suggests this will not have a major impact on the demand for advice in the
foreseeable future. The main implication for polarisation is that it seems likely
to reduce the barriers to entry for both providers and advisers and thus to
improve the efficiency of competition in general.
Regulatory trends also impact on the continued relevance of the current

polarisation regime. A trend towards product regulation, minimum
standards and disclosure to some degree reduces the need for strict
intermediary regulation and hence polarisation. Finally, future EU legislation
may make maintenance of the current polarisation regime unsustainable if
foreign competitors can operate in the UK market without being bound by it.
55
London Economics
July 2000
Section 7 Options for change
7 Options for change
The experience of polarisation is that the details of the rules can be very
important for how competition evolves. In thinking about the future, it is
therefore important to set out the proposed scenarios in some detail before
trying to determine the likely market outcomes.
We have developed four core scenarios, which reflect the analysis of the
economic consequences of polarisation, to examine against the baseline of the
present rules. For each scenario we first analyse the impact it would have on
the market (its market outcome), and then go on in Section 8 to look at how it
performs in terms of regulatory cost-benefit analysis.
7.1 Defining and assessing the scenarios
The scenarios do not correspond exactly to those set out in our terms of
reference, but rather are designed to be:
a) feasible, in terms regulatory implementation;
b) substantial enough changes to have an appreciable effect on the market,

and thus form a suitable basis for analysis; and
c) likelier to yield a positive balance of benefits over costs than other closely-
related alternatives.
The core scenarios are as follows:
• Redefinition of polarisation for all products. In this scenario, firms

could choose to be independent, tied, or multi-tied, with clear
disclosure of both status and interest.
• The DGFT recommendation. In this scenario, collective investment
schemes be removed from the ambit of polarisation.
• Minimum standards exemption. In this scenario, simplified
packaged products (CAT ISAs and stakeholder pensions) would be
removed from the ambit of polarisation.
• Retaining the essentials of polarisation. In this scenario, providers

operating tied distribution are allowed to introduce the products of
other providers to fill gaps in their product range.
In addition to assessing these core scenarios we have reviewed changes along

two other dimensions:
• That within IFA firms a separation be instituted between those who

construct the firmÕs panel of product providers and those who
negotiate commission (one of the DGFTÕs recommendations).
56
London Economics
July 2000
• Permitting a firm to describe itself as ÒindependentÓ only where it

meets certain conditions, namely that it provides advice on a fee for
service basis. In this case a regulatory distinction would be
introduced among members of the existing IFA profession, between
on the one hand fee-based independent advisers and on the other
hand commission-based brokers.
To examine these two variations we have considered how much would

change if they were applied separately within the current polarisation regime.
However, in principle these could be applied to a number of the core
scenarios, either alone or in combination.
In considering how these scenarios will effect the market we have focused on
whether it will bring about a substantial change in the channels through
which investment products are distributed. We do this by examining its
impact on consumers, distributors and providers. This is based on two sets of
considerations and evidence:
a) We first examine the quantitative, documentary and purchaser interview

evidence on how consumers behave today, and how they might change
their behaviour under different scenarios.
b) We then assess whether this behaviour is consistent with the provider

interview and historic evidence on how providers and distributors are
likely to change their behaviour. Ideally, we wish to identify an
ÒequilibriumÓ in which consumers, producers and distributors behave
consistently.
7.2 Scenario 1: Redefinition of polarisation for all

products.
We have defined this scenario in the following manner.
p A new category of ÒFinancial AdviserÓ would be introduced, alongside

IFAs and tied agents. Although, we use ÒFinancial AdviserÓ throughout
this report, there are a number of suggestions for possible names for these
intermediaries. The name that most clearly differentiates them from IFAs
and tied agents should be chosen.
p As today, use of the term ÒIndependent Financial AdviserÓ would be

restricted to:
• advisers who search the whole market for packaged investment

products;
• IFAs owned by providers, who have to apply the Òbetter than

bestÓ rule to packaged products that are manufactured by the
57
London Economics
July 2000
company within the same business group to which the IFA

belongs; and
• advisers who do not have contractual ties for packaged

investment products.
p The rules for companies distributing through tied agents would be

redefined so that they can offer the products of other companies under
the other companyÕs brand alongside their own products.52 They could
do this either by adding products within the same product type, or to fill
a gap in their existing product range. The liability for the advice given by
agents of the provider would rest with one of the providers (the Òlead
providerÓ) chosen by mutual agreement.
p The new ÒFinancial AdvisersÓ would be able to tie to a number of product

providers. The liability for the advice they provide would rest with the
ÒFinancial AdviserÓ.
p Any product that is advised by either ÒFinancial AdvisersÓ or tied agents

must be Òas good asÓ any other product within the product range.53
In this scenario, advice can be provided by three categories of adviser:

Independent Financial Advisers, ÒFinancial AdvisersÓ and tied agents.
7.2.1 Key findings

p The size of the IFA sector will remain similar to that seen today. This is
because customers have a good understanding of the meaning of
ÒindependenceÓ and place a high value on it. The opportunity for
customer confusion will be minimised by the maintenance of status
disclosure.
p Providers are not likely to buy IFAs in large numbers to create larger tied
distribution networks (though this is a risk that needs to be considered).
On the face of it, they would find it attractive to try to purchase cheap IFA
distribution. However, if in practice they did buy up IFAs this is likely to
reduce the cost advantage substantially. IFAs themselves would choose
not to tie, both because they would anticipate that they would end up
earning less than they receive today and because IFAs appear to have a
preference for providing independent advice, i.e. cultural reasons.
52 It should be noted that the responsibility for the product lies with the product provider. Therefore a
provider, say company A, can offer a product using re-insurance or outsourcing arrangement as today.
This will be branded under company AÕs brand and the responsibility of the product remains with
company A. Alternatively, they can offer another providerÕs product, say company B, in which case the
responsibility lies with company B.
53 A product group will be defined in terms of a distinct set of products which the firm markets.
58
London Economics
July 2000
p Providers using tied distribution will introduce new products to fill gaps
in their ranges, replace their existing products in some instances. In
addition, this will introduce competitive pressure on tied providers to
improve their Òown labelÓ products.
p Consumers will gain in terms of more product variety and hence better
advice in the tied channel.
p There will be some market entry by new distributors and new product
providers.
7.2.2 Impact on the IFA sector

A key question for every scenario is the size of the independent intermediary
market. Given the success of the IFA channel over the last five years in terms
of its share of new business we need to assess whether consumers using IFAs
or IFAs themselves would change their behaviour.
a) Customer demand would protect IFAs
The impact of redefining polarisation by introducing a third category of

adviser and allowing tied advisers to sell products from multiple providers
will depend fundamentally on how (if at all) consumers using different types
of distribution today would change their behaviour.
After 12 years of polarisation, consumers using IFAs do appear to have a

reasonable understanding of the difference between IFAs and tied advisers,
though a minority of 20% still do not understand the status of their adviser
(see Section 5.1.1).
Furthermore, the independence of IFAs appears to be their most distinctive

characteristic and the most important factor in consumersÕ choice of adviser Ð
even ahead of personal recommendations. According to the consumer
research for this project, 88% of those using an IFA thought independence
was an important characteristic. Asked for the most important characteristics
of an adviser, for those choosing to use an IFA for the first time, the most
popular answers were:54
• that customers feel they will get independent advice (31%);
• the reputation of the adviser they deal with (21%);
• that customers feel the adviser will cover the whole market (19%);
• recommendations from family, friend or work colleague (13%);
54 Continental Research. Respondents could choose more than one answer.
59
London Economics
July 2000
• that customers think this would be the cheapest (11%); and
• IFA status in general (9%).
It is also useful to differentiate between customers who value IFAs because

they:
• are independent from contractual ties, and work on behalf of the

consumer; and
• offer a wide choice of different providersÕ products.
For some customers, width of search is an important attribute. About 50% of

those using IFAs believe they survey the whole market when selecting a
product suitable for them, rather than selecting from a small number of
companies. Many of these would consider using an alternative channel if
IFAs chose from only a small group of providers.
Figure 20: How would it affect your likelihood of using an IFA, if they
chose from a narrow range of companies?
100%
90% More likely to have used
80% an IFA
70%
60%
Would make no difference
50%
40%
30%
20% A little more likely to have
10% used another way of
buying
0%
t
st
d
on
A
P
en
A lot more likely to have

ru
on
PE
IS
m
si
tT
tb
n
used another way of

Pe
ni
do
en
buying
tm
En
es
v
In
Source: Continental Research. Base: All saying the IFA looked at the whole market
As shown in Figure 20 (above) around 40% of those who believe IFAs survey
the whole market Ð or around 20% of all IFA customers Ð would consider
using another channel. The great majority would continue to use an IFA
60
London Economics
July 2000
rather than buy direct or through a tied agent. However, in this case the IFA
is still free to choose from different product providers.
If we consider the alternative scenario, where tied agents were to offer

products from a selection of providers, only around a quarter of customers
would then consider them a closer substitute for IFAs. IFAs offering a limited
search thus seem much more attractive to current IFA customers than tied
advisers linked to a number of product providers.
Figure 21: How would it affect your likelihood of using a tied agent, if they
offered the products of several providers?
13%
18% A lot more likely to have bought from a
bank, building society or insurance
company salesman
14% A little more likely
Less likely
55%
Source: Continental Research. Base: All buying through an IFA
Whether IFAs would choose to tie depends on the cost savings from ceasing
to provide independent advice, and how they compare to the likely loss of
revenue from doing so.
Evidence from our interviews with small IFAs suggests that few would
choose to change status in the short term. This is because the importance to
small IFAs of commissions on on-going business makes it unlikely that many
would find it beneficial to tie to an existing provider. If they did, they would
not be able to advise on the on-going services of their clients (unless,
improperly they advised on unnecessary churning of client plans for
commercial reasons55), and would lose an important source of income. Some
55 The possibility of this was suggested in submissions from the IFA sector.
61
London Economics
July 2000
advisers might get round the loss of revenue on ongoing business by setting
up a parallel IFA firm to service it. Others might find it worth becoming
ÒFinancial AdvisersÓ because of the cost savings from not having to search
the whole market and the improved terms offered by the providers to which
they are tied. But these exceptions are likely to be small, and Ð as when
polarisation was first introduced Ð few advisers are likely to change their way
of doing business in the short run.
But whether this is a long-term equilibrium depends on longer-term cost and

revenue considerations. Again, we conclude that there would be little or no
incentive to give up IFA status. Consider the following trade-offs:
• If an IFA were to give up independent status becoming a ÒFinancial

AdviserÓ but not tie to particular providers, it might lose 20% of its
customers and income, representing an income loss of £5,000-£7,500
on average. This represents the number of consumers who currently
use an IFA but who would consider using other channels if the IFA
restricted their search (discussed above). The likely revenue
implications of this are far more than the £1000-2000 per annum
needed to buy in market-wide search data from an information
provider.56
• If an IFA were to give up independent status and tie to a few

providers, it might lose up to 75% of its business to those remaining
as IFAs (derived from the proportion of consumers who would
consider a tied agent offering multiple products a close substitute to
using an IFA discussed above). This is far more costly than any
possible saving from reduced training and competence costs. IFAs
can already cut their T&C costs by becoming members of networks,
but many choose not to, suggesting that these costs do not outweigh
even that loss of autonomy.
Being based on stated preference57 data this evidence is not conclusive.

However, it is broadly supported by our interviews with a range of
independent intermediaries. In particular, we found support for the value
consumers place on independence, and a reluctance to provide tied advice on
the part of both national IFAs and small IFAs, (the evidence from the IFA
networks was inconclusive).
b) Providers would not find it profitable to buy up IFAs
As shown in Figure 22, IFAs offer a considerably cheaper method of

distribution than either direct sales forces or appointed representatives.
56 This represents the cost of a restricted information service. The cost can be as high as £20,000 for more
comprehensive services.
57 Stated preference is where consumers are asked hypothetically what they would do. Revealed
preference is where observation about how consumers behave reveals their preferences.
62
London Economics
July 2000
Figure 22 Costs as a percentage of new business, 1997
100%
90%
80%
Costs as a % of APE
70%
60%
50%
40%
30%
20%
10%
0%
Direct Salesforce Appointed IFA Bancassurer
Representative
Source: Insurance Pocket Book 1999, Tillinghast Towers Perrin
This cost advantage is thought to be a mixture of:
• lower retention costs (IFAs do not need to make significant outlays to

retain their customers compared to insurance companies as
illustrated in Figure 15);
• higher retention rates (evidence from our interviews with national

IFAs); and
• higher premium amounts58 (see Section 4.1.2).
Even allowing for premium size there is evidence that IFAs are a cheaper
method of distribution.59 It is possible that providers, knowing this cost
advantage, will try to take advantage of the redefinition of polarisation to buy
up IFAs.
However, our previous reasoning suggests that tying to product providers is

likely to reduce efficiency quite severely, because consumers would tend to
58 This is supported by evidence from national IFAs and small IFAs in our provider interviews.
59 ÒPerformance benchmarking in financial services: international evidence from the life insurance
industryÓ, Paul J. M. Klumpes, 30 September, 1999.
63
London Economics
July 2000
abandon ÒFinancial AdvisersÓ in favour of genuinely independent advisers.

Therefore in our view:
• it is unlikely that product providers will find it worthwhile in the

long run to buy IFAs (it is worth noting that many of the providers
who would be in a position to ÒbuyÓ distribution have in fact cut
back significantly on the size of their direct sales force in recent years
and that such a policy would represent a considerable reversal of
strategy); and
• it is unlikely that IFAs will be willing to be bought, as they would

correctly surmise they may well end up earning the returns of todayÕs
tied agents (see Table 6 below), which are well below those of IFAs.
Table 6: Remuneration of advisers
IFA Tied
Entry £20,000 £16,000
After five years £30,000 £25,000

Source: Recruitment consultancies
Evidence from a recent survey of IFAs also suggests that small IFAs are not
greatly concerned by the introduction of multi-ties. While 60% of small
directly regulated IFAs considered training and compliance to be a major
threat, less than 20% thought the introduction of multi-ties posed a greater
threat.60
Finally, evidence from IFA associations suggests there are strong cultural
reasons why IFAs will choose to remain independent. Many IFAs were
previously tied agents and will not wish to return to this status.
In summary, although some providers might attempt to buy IFAs, it is

unlikely that the IFA sector will be substantially reduced. Few IFAs would
become multi-tied ÒFinancial AdvisersÓ or tie to a single provider.
It remains possible that providers will buy up distribution in the short term,
and this is a risk worth considering. It would however only affect the long
run equilibrium if it were irreversible, and there is little evidence to support
this idea. First, experience over the last ten years suggests that providers
have adopted different strategies on distribution but where these have
proved misguided they have been reversed. Second, there is considerable
scope for reversal because of the relatively free movement of individuals
between the tied and independent sectors.
60 Evidence from a survey undertaken by Equus Alliance reported in Money Marketing April 13th, 2000.
64
London Economics
July 2000
Another reason why companies might not engage in a distribution war, even
in a transitional period, is that they cannot afford one. Life insurance
companies in recent years have faced an £11 billion mis-selling bill, an
estimated £15 billion guaranteed annuities bill and now face the prospect of
the high capital strain of writing stakeholder pension business as well as
development of new e-commerce and IT systems. Indeed, life analyst Ned
Cazalet published a report last year61 which found free cash in the life
insurance sector was at the lowest level in more than 10 years. Firms are
under great pressure to cut costs, including business acquisition costs.
Australia provides a useful case study to examine the implications of

allowing multi-tied advisers whilst attempting to protect the status of
independent agents. Financial planners there cannot describe themselves as
independent if they have:
• ownership links with a financial services institution;
• restrictions on the products they offer;
• relationships with institutions through commissions or incentives;
• any other links which could influence their recommendations.
The evidence on the face of it is not particularly reassuring. Of the top 10

groups in the financial advice market, only one is not owned by a large
financial institution. These companies, as in the US, offer a mixture of
proprietary and non-proprietary products, for investment funds in particular.
In addition, there is evidence that the larger institutions have been buying up
distribution for their investment products.
However, there are a number of vital differences between the Australian

situation and our scenario:
• The independent intermediary channel and the meaning of

ÒindependenceÓ is more firmly established in the UK.
• The scenarios do not exclude commission payments as a form of

remuneration of the independent channel, whereas the Australian
system is closer to our Variant B where there are restrictions on
independent advisers receiving commission.
• The market for life and pensions and unit trust is much less
concentrated in the UK, making the value of independent advice,
higher here than in Australia.
61 Cazalet Financial Consulting, WolfÕs At The Door, Sept. 1999. http://www.cazalet-financial.com.
65
London Economics
July 2000
7.2.3 Impact on the tied distribution channel

Redefining polarisation would allow tied providers to offer the products of
other providers:
• to fill a gap in their current product range;
• to replace a product in their range; and/or
• as a competing offering alongside a product they already provide.
Gap-filling. This will increase both the volume of business undertaken and
the quality of advice in the tied channel (e.g. because it allows a more suitable
product to be sold). It is likely to happen where:
• Producing the product oneself is prohibitively expensive because of

lack of expertise or capital. Gap filling products can already be
provided through re-insurance and out-sourcing arrangements and
have been where their value is determined to be high enough.
• The product is provided by niche providers such as friendly societies

who are currently unable to provide their products through tied
distribution networks.
• There are innovative products that would traditionally have taken

some considerable time to replicate. Evidence was presented to us
from providers offering specialist protection products and off-shore
products that have not been developed in the tied channel.
• The distributor is small and has been prevented from providing

product by the cost of providing advice across the whole marketing
group. Redefinition would enable them to tie to a number of
providers and undertake the training and competence commitment
themselves. This may prove profitable as they must provide advice
only over the limited product range to which they are tied.
• There are very significant economies of scale. This is particularly

relevant for stakeholder pensions.
Replacement. Replacing oneÕs own product with that of another company is

most likely to occur where the other company has a stronger brand or there
are gains to be made through lower cost (i.e. as re-insurance costs or systems
costs can be avoided).
Offering competing products of the same type. Few providers currently use
multiple brands. It is only allowed within marketing groups, and providers
argue that it is expensive, as advisers must advise on the entire range of all
providers in the marketing group. It is also difficult because the advice
generally comes down to recommending products with the lowest charges (as
within the marketing group there are few other differentiating factors).
66
London Economics
July 2000
Changing the rules so that the adviser needs to give best advice but only
within the chosen product range would increase the attractiveness of
introducing multiple brands. This was supported by a number of banks and
building societies in the provider interviews.
This will lead to some increase in training and competence costs Ð staff need
to be trained on multiple products and need to justify advising on particular
products. (If companies choose to do this it must mean that the additional
revenues outweigh the additional costs.)
The most likely multiple offerings are unit trust/ISA products, which have
larger consumer brands and relatively small training costs. The growth of
fund supermarkets, and the trend towards multi-manager products, suggests
that multiple offerings do have business advantages. This is supported by
evidence in the USA, where providers can and do distribute their own
products and those of other providers. The split between proprietary and
non-proprietary varies significantly between providers but the latter can be
substantial (see Table 7).
Table 7: Percentage of proprietary and non-proprietary funds sold by the

top five firms in the USA in 1998
Proprietary Funds Non-proprietary Funds
PaineWebber 10% 90%
Salomon Smith Barney* 25% 75%
Prudential Securities* 25% 75%
Merrill Lynch 45% 55%
Morgan Stanley Dean Witter 80% 20%

Source: Does Practice Contradict Spirit of Tully Report?, On Wall Street, September 1999. * - To date for
1999 these companies estimate their proprietary/non-proprietary sales mix at 35%/65%.
A perverse detrimental effect could occur if providers were to use the ÒhaloÓ
effect of having a strong branded product in their range to sell more of their
own product. But it is unlikely that the brand owner would accept this, or
that weakly-branded tied providers would wish to introduce a strong brand
alongside their own.
Allowing multi-ties will probably result in distributors Ð particularly private

banks Ð entering the market offering their own products alongside those of
other companies. This service would probably be mainly aimed at high net
worth customers: we were told in industry interviews that when firms use
this model in the USA, they often offer non-proprietary products only to
customers with assets above a certain level.
67
London Economics
July 2000
Redefinition also allows new providers to enter using both the tied channel
and the IFA channel, but we feel it will not do much to lower barriers to this
type of entry. IFAs will still demand that providers be financially secure and
able to show good past performance, and it will still be as difficult for new
providers to distribute through them. Banks are already allowed to distribute
the products of efficient manufacturers through outsourcing and re-insurance
deals. Unless the new provider already has a strong customer brand, it is not
clear why banks (or other tied distribution networks) would start promoting
it.
7.2.4 Competition and consumer detriment

The effect on competition would nonetheless be positive. Competition
through the tied sector providing other companiesÕ products is likely to lead
to lower charges and greater conversion rates. In addition, the quality of
advice would rise as it is more likely that a suitable product would be
available.
It is more difficult to gauge the overall effect on the number of providers in

the market in the long run and their degree of specialisation. The polarisation
rules have distorted the structure of the market, by encouraging providers to
offer as many products as possible and creating more incentives for in-house
provision than for outsourcing. Removal of such fundamental distortions is
an important consideration for the long-run competitiveness of the UK
financial services industry.
Redefining polarisation increases the potential for confusion over status.

There are two ways in which consumers can be confused. The first is where
independent advisers act as if they are multi-tied. The second is where the
multi-tied or tied agent portrays themselves as independent.
The regulatory definitions of independence in this scenario are as at present

so that the first type of masquerading is no more likely than at present. The
second type of masquerading is to some degree already possible from a tied
adviser selling products with multiple fund choices. As this maintains the
distinction between an independent adviser and other types of adviser, the
potential for masquerading is somewhat limited.
However, status disclosure may need to be strengthened. For example: agents

could be required to carry a business card which states: ÒTied financial
adviser marketing only the products of Company A and Company B. Only
an independent financial adviser advises on the most suitable products from
all companiesÓ. As in the US, agents could be required to disclose the
percentage of business written by each firm.
Status disclosure should assure that consumers know to go back to their

adviser for issues surrounding advice, and to the product provider for issues
surrounding the product. However, it must be acknowledged that
68
London Economics
July 2000
introducing the ÒFinancial AdviserÓ category will somewhat reduce

consumer clarity about adviser status.
7.3 Scenario 2: The DGFT recommendation
Following the DGFTÕs recommendation set out in the OFT report, this
scenario is defined by the following rule changes:
• collective investment schemes would be removed from the

polarisation rules altogether; and
• advice on any providerÕs collective investment scheme products

could be provided by any agent or adviser (as is currently the case
with non-polarised products, such as mortgages).
In this scenario, advice on the remaining polarised products (i.e. life and
pensions products) can as at present only be provided by Independent
Financial Advisers and tied agents.
7.3.1 Key findings

p Removing collective investment schemes from polarisation may enhance
competition between collective investment schemes but may be
detrimental to competition between collective investment schemes and
other products.
p It is possible that this will increase the potential for consumer detriment
through increased confusion over the status of the adviser.
Intermediaries will be able to advise customers about substitutable
products in different capacities and tied advisers may be able to
masquerade as independent.
p There is considerable doubt over whether collective investment schemes

represents the most appropriate product boundary.

The effect of this scenario is to allow IFAs to tie to a number of providers of
collective investment schemes, and to allow tied distributors to offer the
collective investment schemes of different providers. The argument in favour
of the DGFT proposal is that these products are simple, so that:
• consumers can shop around, which is evidenced by a higher

percentage buying direct (as illustrated in Figure 2); and
• there is less consumer detriment than for more complex products

(this is supported by evidence presented in Section 5.1.1).
69
London Economics
July 2000
However, there is still considerable doubt about whether competition will

provide sufficient protection for all types of customer. Many consumers do
not perceive collective investment schemes as simple. As shown (Figure 23)
even though customers are less likely to seek advice on unit trusts than on
other products, well over 40% of consumers do seek advice when purchasing
unit trusts.
Figure 23: Demand for advice, by product
100%
90%
80%
70%
60% No - just arrange the
transaction
50%
Yes - did seek advice
40%
30%
20%
10%
0%
st
P
n
A
d
y
io
PE
ru
on
IS
ic
ns
ol
tT
tb
tp
pe
ni
en
en
U
al
m
on
st
w
ve
rs
do
Pe
In
En
Moreover, according to an FSA survey, financial decision-makers find unit

trusts and ISAs more complicated than pensions.62 This throws some doubt
on the argument that trusts are simple enough to be removed from
polarisation.63
What would be the impact on customers of taking collective investment

schemes outside the polarisation rules? At first glance, it would not lead to a
very large reduction in the IFA sector.
62 Of those who were familiar with products, 51% believed pensions were straightforward as opposed to
only 39% for unit trusts, 46% for cash ISAs and 35% for equity ISAs. Better informed consumers, FSA
63 One note of caution on this is that some products, such as pensions, may be considered to be
straightforward because consumers do what advisers suggest, rather than because of their inherent
features.
70
London Economics
July 2000
Our consumer research suggests that only around a quarter of those currently
purchasing unit trusts and ISAs through the IFA channel would be more
likely to use banks, building societies or insurance companies if they offered
the products of more than one provider (Figure 24).
Figure 24: Response of IFA customers, if tied providers were to offer

multiple unit trust products
11% 10%
A lot more likely to have bought from
a bank, building society or insurance
company
16%
A little more likely
Less likely
63%
Source: Continental Research. Base: All using an IFA to buy a unit trust/PEP/ISA.
There could also be a knock-on effect on the attractiveness of using IFAs for
products that are still polarised. But again only about 27% of people buying a
life and pension policy through an IFA would consider other channels if their
advisers were to limit themselves to a small number of unit trusts.
It might however produce more important changes in the competitive

advantage of different types of provider. In particular, it would enhance the
proposition of fund managers offering unit trusts, and weaken that of
providers of unit-linked life policies. The result of this would be an increase
in the sale of unit trust products. As unit trust charges are not significantly
lower (nor significantly less variable) than those on competing products,
biasing products in their favour is not obviously beneficial.64
The extent of the detriment will depend on the degree of substitution

between unit trusts and other products. This is likely to vary with customer
64 PIA Disclosure report 1999.
71
London Economics
July 2000
characteristics Ð for those in lower income strata, pensions and unit trusts
may be very similar savings vehicles65, and accordingly more substitutable
than they are for other groups.
Just how close the substitution could be is shown by the example of the ISA
market. While unit trust ISAs would be outside the polarisation regime, ISAs
with a life component and a unit trust component would be inside. IFAs
would have an incentive to sell ISAs without a life component and the rules
for investment advice would differ for products within the same category.

Tied advisers will be able to offer collective investment products from more
than one provider, and are likely to compete by offering a range of unit trusts.
To the extent that these products are simple, this might encourage increased
price competition. In particular, it would allow fund managers to distribute
their products both through tied channels and IFAs (though some of the
benefits of this have already been exploited through the outsourcing of fund
management and multi-manager products).
Tied advisers may therefore benefit from greater customer demand for their
services, both in collective investments, and to a smaller extent, with cross
selling, in life and pensions. This should slow or reverse the trend in the
number of tied advisers.
Other product providers would be able to compete with IFAs on collective

investment products by offering, for example, a multi-tie using the US model
of a number of unit trust funds. With the formation of multi-ties, IFAs might
move to differentiate themselves by offering a greater range of products on
their panels. This would have beneficial effects in terms of offering consumers
a larger variety of products.

Detriment could arise because the adviser will be advising on both polarised
and non-polarised products that potentially meet the same need. This could
cause confusion of two kinds:
• over the incentives of the adviser to sell a particular product; and
• over the status of the adviser in general.
A tied adviser could for example say that he was independent for collective
investment products, but tied for life and pensions products. Equally, an IFA
65 See M Cook and P Johnson ÒSaving for retirement: How taxes and charges affect choiceÓ May 2000.
FSA Occasional Paper 8.
72
London Economics
July 2000
will be able to claim to be independent for polarised products but will not
need to be for collective investments.
This problem is illustrated by evidence from the mortgage market. In this

case, mortgages and life products (which are complementary products rather
than substitute products) are often sold together. However, an independent
adviser can offer only a limited range of mortgage products and a tied
adviser can offer independent advice on mortgages. Consumer Panel
research suggests this has lead to consumers being confused over the status of
their adviser and wishing they had used an alternative type of adviser66.
We can therefore see that there are a number of potential problems with this
scenario in terms of consumer detriment and the potential effectiveness of a
regulatory change of this nature.
7.4 Scenario 3: Minimum standards exemption
This scenario is defined by the following rules:
• simplified products are removed from the polarisation rules

altogether;
• advice on any providerÕs simplified products can be provided by any

agent or adviser; and
• simplified products are defined as those currently polarised products

with standardised product terms set by the regulatory authorities.
Currently this would apply to CAT-standard ISAs and to stakeholder
pensions, which have to meet standards set by the Treasury.
In this scenario, advice on the remaining polarised products (i.e. currently

collective investment schemes other than CAT-standard ISAs and life and
pension products other than stakeholder pensions) can as at present only be
provided by Independent Financial Advisers and tied agents.
7.4.1 Key findings

p The effect on consumer detriment is likely to be small. Consumers who
buy a simplified product in the mistaken belief that the adviser is
independent will be protected by the minimum standards. Nor will tied
providers get much of an independence ÒhaloÓ from selling simplified
products Ð their halo comes from Government, not providers.
66 Financial Services Consumer Panel response to: HM Treasury Discussion on Mortgage Regulation,
October 11, 1999.
73
London Economics
July 2000
p It will however not bring about much of an increase in the distribution of

simplified products, as they only offer small distribution margins. But it
may increase their availability by making economies of scale possible for
product providers.
p The competition advantages are also likely to be small, as these products

are already tightly priced.

IFAs currently sell very few CAT-marked ISAs, and it is predicted that they
will also sell relatively few stakeholder pensions (directly to consumers at
least). This is because it is likely to be uneconomic for them to do so at the
levels of remuneration available, and does not reflect commission bias. Being
able to tie to a single provider for these products will cut costs and may
increase the number of IFAs offering them. But given IFAsÕ low sales
expectations for these products, it is not clear that their removal from the
polarisation regime will make any significant difference.
Our industry interviews revealed that IFAs may start offering simplified
products on a fee basis, as part of a general trend for consumers to consult
their adviser about simple products before buying them direct. If IFAs were
to introduce a fee, it would probably increase the proportion of ISAs bought
direct.
There is some concern that IFAs who tie for simplified products will still be
able to use the title ÒindependentÓ (because they fulfil the criteria for
independence on polarised products). It may therefore be the case that some
IFAs use the ÒindependentÓ title, but sell mainly non-polarised products for
which they are in fact tied. This seems unlikely to be a very profitable
strategy, because of the tight price cap on these products. If it is of concern, a
regulatory limit could be placed on the amount of tied business agents can
conduct while describing themselves as ÒindependentÓ.

Tied agents would be able to offer the simple products of more than one
company. However, the charge limits on simplified products would probably
make this unattractive. Few providers are now choosing to offer their own
CAT marked ISAs, and it is unlikely that they would choose to offer a variety
of other standardised products either.
The main counter argument to this is that, if there are considerable economies
of scale in standardised products, distributors may be able to earn a higher
commission by selling someone elseÕs popular product rather than the small-
scale product of the provider they are tied to. This may increase tied
distribution of these products, which could be very important for stakeholder
pensions.
74
London Economics
July 2000

The effect on consumer detriment and competition (through entry and
innovation) are likely to be very modest, due to the standardisation imposed
on the design of simplified products.
7 . 5 Scenario 4: Retaining the essentials of

polarisation
This scenario is defined by the following rules.
p As today, use of the term ÒIndependent Financial AdviserÓ would be

restricted to:
• advisers who search the whole market for packaged products;
• IFAs owned by providers, who have to apply the Òbetter than

bestÓ rule to packaged products that are manufactured by the
company within the same business group to which the IFA
belongs; and
• advisers who do not have contractual ties for packaged

investment products.
p Rules on providers distributing through tied agents are redefined such

that providers can offer the products of other companies under the other
companyÕs brand, but only to fill a gap in their existing product range or
to replace their own product. The liability for advice is their responsibility
as the lead provider.67
In this scenario, advice on polarised products could as at present be provided

only by Independent Financial Advisers and tied agents.
7.5.1 Key findings

p This is less likely to lead to consumer confusion over adviser status than
the redefinition scenario, as tied advisers can still only sell one product
from each provider and they cannot promise choice. However, they will
be able to claim they have chosen the ÔbestÕ product range, potentially
adding to consumer confusion.
67 The responsibility for the product remains that of the product provider. The brand under which the
product is marketed therefore represents where a consumer should go if they have queries regarding
the product. Separating the brand of the product and the provider who has responsibility of product
can only lead to customer confusion.
75
London Economics
July 2000
p Multiple brands would, however, be much more complex than those on

the market today where providers offer different brands for insurance
products and mortgage products but not for polarised products.
p Consumers would gain from being offered a complete range of products

by tied distributors.

People who receive independent advice would continue to be attracted by the
fact that IFAs search the whole market, are not tied to particular providers
and give independent advice.
Existing small or large IFAs would be unlikely to tie to providers, as the tied
advisers will be very similar to what they are today. There would be
therefore be little change in the current IFA numbers. New consumers
entering the financial services market, however, could be influenced by the
greater selection of brands the tied provider could offer.

Allowing gap filling would affect a number of firmsÕ offerings, enabling them
to place new products in their range. The advantages from gap-filling would
be similar in kind to those outlined in the redefinition scenario.
The products introduced are likely to be relatively less important products as,
if their value had been substantial, they would already have been introduced
through out-sourcing or re-insurance deals.
Gap-filling will save tied providers the current costs of re-insurance and
additional system costs, enabling them to cut the charges on these products,
benefiting consumers. However in cases where they offer a new product they
will assume additional training and compliance costs.
There would be new entry by product providers who:
• fill a gap in someone elseÕs range; and
• provide a product with their own brand whereas to date they have
been providing out-sourcing or re-insurance services.
In some cases, providers will continue to use outsourcing and re-insurance as

they believe their own brand is stronger than that of the other providers.
By offering slightly more choice this is likely to increase the cross-selling of

existing tied providers.
76
London Economics
July 2000

Consumer detriment would decrease in the tied channel, as advisers become
able to offer gap-filling products that are more suitable to customer needs.
Consumers may also benefit from the cost savings providers achieve through
not having to work around polarisation rules using reinsurance. The IFA
sector will be largely unaffected, so there should be no change in consumer
detriment through that channel.
7 . 6 Variant A: Panel/commission separation

within IFA firms
The DGFT recommended that the selection of products for IFAsÕ product
panels or recommended lists should be formally separated from the
negotiation of commission on these products.
Our interview evidence confirms that a number of industry participants,

including both national and network IFAs, already make this separation, in
line with a 1997 PIA Consumer Panel report giving good practice guidance on
product panels. But it is not clear how far smaller IFAs can make this
separation. It is also clear that there are limits to the effect that separation can
or should have on IFAs: some productsÕ commission levels may be below the
level at which an IFA can profitably run his business, and sourcing them onto
panels may not be worthwhile.
Other PIA recommendations of good practice are:
• clear, open and objective criteria for inclusion, so the process is

transparent, with a particular focus on customer benefits;
• no set number of product providers, with all those meeting the

criteria being included;
• adequate research resources and product reviews at least yearly;
• provision for advisers to go Òoff-panelÓ on reasonable grounds

without unjustified obstacles being placed in their way; and
• clients to be clearly informed of the panelÕs operation and objective.68
We see these safeguards as already quite substantial and hence do not at

present see an economic case for expanding them.
68 PIA Consumer Panel Report 1997.
77
London Economics
July 2000
7 . 7 Variant B: Only fee-based advice can be

ÒindependentÓ
The idea motivating this variance is the suspicion that commission-based

advice is bound to be biased, whatever safeguards are put in place, and that
true independence can only be achieved by those who provide advice on a fee
for service basis. Regulation would distinguish between IFAs who provide
advice on a commission basis (Òcommission brokersÓ) and those providing
advice based on fees (ÒIndependent Fee-based AdvisersÓ). Fee based advice
could be provided on an hourly rate or as a fixed charge.
The effect would be to increase the transparency of the price paid by

individuals for advice. This is likely to impact different types of customers:
• those who value the term ÒindependenceÓ but do not understand its
implications, i.e. that independent advisers are currently largely
remunerated on commission;
• those who do not understand how much they currently pay for
advice; and
• those not currently using an IFA because they do not believe they are
truly independent.
However, we have not considered all the possible alternatives. In particular,

payment on commission has two dimensions:
• conditionality on purchase; and
• neutrality between providers of the same products.
Other alternatives which combine payment conditional on purchase with

neutrality for advisers could be beneficial and should in our view be
considered in the future, perhaps as part of a broader review of all
compensation and remuneration for packaged products.

Our industry interviews suggested that many or all IFAs offer all their
consumers the choice between fees and commission. However, the evidence
of consumer research (Figure 25) suggests either that it is not the case or that
it is the case but only 20-30% of customers remember being offered this
choice. It is also known that only about 10% of business is conducted on a fee
basis. Surprisingly, this does not appear to vary by product purchased.
78
London Economics
July 2000
Figure 25: Were you offered a choice of fees or commission?
100%
90%
80%
70%
60%
DK/can't remember
50% No
Yes
40%
30%
20%
10%
0%
Personal pension Endowment Investment bond Unit Trust PEP ISA
policy
Source: Continental Research. Base: All buying through an IFA.
If few consumers understand that they are being given a choice between fees
and commission, the low current take-up of fees does not necessarily mean
that consumers would not consider going to a fee for service adviser if this
was the only way to get ÒindependentÓ advice. To find this out, one has to
ask consumers how they would feel about this hypothetical position. Our
survey research shows only half of recent purchasers would consider paying
fees at any level. Of those who would consider fees, the amount they are
willing to pay is set out in Figure 26 below.
79
London Economics
July 2000
Figure 26: How much would you be willing to pay for advice?
11%
2%
1%
2%
Up to £100
13% £101-£200
£201-300
£301-£400
£401-£500
Over £500
71%
Source: Continental Research. Base: All who would pay a fee rather than commission if offered a choice.
Providing advice on personal investments on a face-to-face basis generally

takes about 2 hours, and the average hourly rate for an IFA is £90 an hour,
suggesting that the average fee would need to be around £180. This is well
above what most customers say they would be willing to pay.69 It is thus
unsurprising that consumers who are offered fee-based advice do not choose
to take it.
New technology may change this, as has been shown by the recent entry of
the online advisers. They provide advice over the Internet for a fee of less
than £100 (though the administration associated with the purchase is paid for
separately). There is, however, substantial doubt over whether consumers
will accept advice over the Internet or will always prefer face to face contact.
To determine whether fee based advice should be promoted by regulation we

need to determine the reasons behind consumersÕ lack of interest in it. There
are (at least two) competing hypotheses:
• consumers take a rational decision on how they wish to pay for

advice; or
69 This result also throws considerable doubt on whether changing the tax status of fee-based advice
would have any significant effect on the level of demand.
80
London Economics
July 2000
• consumers mistakenly assume there is no commission bias.
In favour of this being a rational phenomenon, it is argued that consumers

understand that commission means they only pay for the advice if they
choose to buy. As they are not sure in advance whether they want to make a
purchase, it is best for them to wait and see what the advice is before deciding
whether to pay for it. In this case, consumers prefer a commission based
system and differentiating between advisers is unlikely to change demand.
The alternative hypothesis is supported by the fact that only 16% of IFA
consumers believe that the product selection was based on the level of
commission. This compares with 30-40% of those using other channels who
believe IFAs choose product on this basis. This suggests that fear of
commission bias might be a reason why so many people do not use IFAs
today.
If the problems of commission are indeed more significant than consumers

believe, it could be beneficial to promote fee based advice. But we do not
believe the existing evidence suggests that IFA commission bias is a serious
issue.
Evidence for commission bias
The opportunity for abuse certainly exists, as IFA consumers largely trust
their advisers and take their recommendations (see Section 5.1.1). Three
arguments are often used to suggest that commission bias is serious:
• Provider bias: it is often argued that providers offering higher

commission will ÒbuyÓ market share. We did not find evidence to
support this, however. Although the leading providers with
significant brands could pay below-average commission, we did not
find that changes in commission had a significant effect on market
share.70
• Payment type: it is often suggested that the trend from regular

premium products to single premium products might be put down to
differences in commission. This is contradicted by the fact that the
commission on regular premium is normally larger than the single
premium equivalent.
• Product bias: finally, it is suggested that differences in commission

could be the explanation for the growth in the investment bond
market relative to the unit trust market. However, the RIY in both
products appears to be very similar, suggesting that at least on this
measure the impact on consumers is limited. Commission
70 This analysis was undertaken using commission rates and new business statistics based on the
disclosure reports. This excluded differential commission levels negotiated with the IFA networks.
81
London Economics
July 2000
differences would therefore appear to be a reflection of how returns

are divided between manufacturer and distributor rather than a
source of detriment to the consumer.
Despite anecdotal evidence that some IFAs and IFA networks do take
advantage of their position to recommend product that yield them the
greatest commission, there is little sign that this is happening on a large scale.
This view was supported by our interviews with a wide selection of
providers using the IFA channel.
In summary, we do not believe that changing status disclosure to reflect the

differences in remuneration would materially change consumer behaviour or
the size of the fee based market.
82
London Economics
July 2000
Section 8 Cost-benefit analysis
8 Cost-benefit analysis
The FSA subjects policy options to Cost Benefit Analysis (CBA) in order to
accomplish its stated objective of cost-effective regulation.71 In addition,
clause 150 of the Financial Services and Markets Act imposes a statutory
commitment to conduct CBA on the FSA.
The purpose of CBA is to try to assess, in quantitative terms where possible,
and in qualitative terms where it is not, the economic costs and benefits of a
proposed new policy. Any CBA of a new regulatory proposal seeks to
compare a proposed new regulatory environment with the old. CBA offers a
broad, yet disciplined, framework within which to weigh up the policy
options.
The FSA uses a CBA framework that classifies costs and benefits under the
following headings:
• direct costs to the regulator;
• compliance costs for the providers/advisers;
• effect on quantity of product sold;
• effect on the quality of products;
• effect on variety of products available; and
• effect on the efficiency of competition.
This framework takes into account the welfare of all the market participants Ð
providers, advisers, consumers and the regulator Ð before drawing policy
conclusions. This is the CBA framework used to derive the summary tables in
Section 8.1 and detailed in the appendix.
The conclusions that are drawn from assessing a scenario using the CBA
framework can in principle differ from the conclusions drawn from assessing
the scenarios on the basis of how far they promote the FSAÕs objectives. In
most circumstances however the two assessment frameworks are likely to
give similar results. This is because:
• A policy change which has an adverse impact on the FSAÕs objectives

usually shows up as a cost in the CBA framework. E.g. a policy
change which harms market confidence is likely to reduce quantity
bought and sold.
• A policy which promotes the FSAÕs objectives usually shows up as a

benefit in the CBA framework. E.g. a policy change which promotes
71 ÒPractical Cost-Benefit Analysis for Financial Regulators: Economics of Financial RegulationÓ, Central
Policy unit, FSA.
83
London Economics
July 2000
consumer protection is likely to lead to quantity bought and sold

rising.
However, cost-benefit results do not imply recommendations for the course

of action the FSA should take.
8.1 CBA summary tables
The incremental costs of the scenarios are shown in Table 8. The DGFT
proposals have far the highest costs, though those of redefinition are not
insignificant.
Table 8: Significant incremental costs
Redefinition DGFT Minimum Retention

standards
exemption
Direct costs One-off - £1 mn One-off - £40k One-off - £40k One-off - £40k

Annual - £3 mn
Compliance One-off - £12 mn

costs Annual - £18 mn
Competition Annual - £56 mn
Quantity
Quality Annual - £48 mn
Variety
Table 9 summaries the benefits of the scenarios described in Section 7. The

redefinition scenario offers the most significant benefits in several categories
(quality, quantity and variety in particular). In our view, these benefits
sufficiently outweigh the cost of redefinition to make it the most economically
attractive scenario. The minimum standards exemption and retention
scenarios, with some benefits and insignificant costs, are also economically
attractive. Meanwhile, it is not clear whether the benefits of the DGFT
scenarios would necessarily outweigh its cost.
84
London Economics
July 2000
Table 9: Significant incremental benefits
Redefinition DGFT Minimum Retention

standards
exemption
Compliance costs ** * **
Competition *** *** * **
Quantity **** *** ** ***
Quality *** * * **
Variety **** ** * **
85
London Economics
July 2000
Section 9 Appendix: Detailed CBA
9 Appendix: Detailed CBA
9 . 1 Scenario 1: Redefinition polarisation for all

products
Direct costs for the FSA
Cost: IFAs who choose to become financial advisers would need to register
with the FSA. This is a one-off cost. As they were previously operating as
IFAs the incremental cost is unlikely to be very significant. However, we
estimate the upper bound on this cost using the cost of re-authorisation. The
FSA estimates the cost of authorising each IFA to be £900.72 At present there
are approximately 25,000 IFAs in the UK73, but only a few are expected to
choose to become FAs. If around 5% were to convert, the FSA would need to
authorise 1,200 FAs at a cost of approximately £1 million.
Cost: The FSA would have to supervise Financial Advisers. This is an

ongoing cost. On the basis of how much it costs the FSA to supervise a one-
person IFA firm74, it appears that it will cost the FSA about £400,000 per
annum to supervise the 1,200 FAs.75 However, as they were previously IFAs
this is unlikely to represent an additional cost.
Cost: The FSA would have to rewrite polarisation rules. This is a one-off cost.
The FSA estimates that this cost will amount to approximately £40,000. This
cost appears to be invariant to the scenarios considered.76
Cost: The FSA would have to ensure that tied advisers adhere to the status
disclosure and Òas good as any other product within the product rangeÓ
obligation. This is an ongoing item, and it does represent an additional cost as
few tied providers currently offer multiple products in the same product
72 Authorising an IFA involves about a day of an FSA AssociateÕs time and 3 days of an FSA
AdministratorÕs time. A day of an AssociateÕs time costs £300 and a day of an AdministratorÕs time
costs £200 (based on daily salary plus overheads). The cost to the FSA of authorising an IFA is £300+(3
x £200) = £900.
73 Source: section 4.3.1.
74 Costs of supervising a one-person IFA firm: 2.2 days of an AssociateÕs time (1 day on site, 0.7 day
preparing report, 0.5 day responding to report; source: David Stuart, FSA). If this takes place every two
years, this represents an annual cost of 1.1 days of an FSA AssociateÕs time, and an annual cost of £330.
75 1,200 x £330 = £400,000 approximately.
76 The FSA expects to expend around 100 person days on rewriting the polarisation rules. Based on the
figures in the FSAÕs 1999/2000 budget, the cost of an average FSA person day has been estimated to be
£400. By implication the cost of 100 person days is estimated to be £40,000.
86
London Economics
July 2000
category. This therefore is an increased burden compared to the current

suitability requirements. If we assume a large proportion (60%) of tied
advisers sell the products of competing providers and one quarter of the cost
of monitoring an IFA is applicable to provider choice within a product
category. This represents a cost of £3 million a year77.
Compliance costs for the industry
Cost: Providers would have to ensure that tied advisers comply with the Òas
good as any other product within the product rangeÓ obligation across a
wider range of products. This is an ongoing cost. An upper bound can be put
on this cost in the following way:
Assume that the ongoing compliance costs of monitoring sales of different

providers in the same product category represent 25% of the costs of
compliance for an IFA, which the FSA estimates at £2,000 per annum.78 There
are approximately 60,000 tied advisers. On the assumption that 60% of
current tied advisers would advise on multiple providers, the ongoing
compliance cost would be £18 million per year79.
Cost: If advisers changing their status are required to notify their customers
this will have an additional costs. It is estimated that this will represent a
one-off cost of approximately £2 million80.
Cost: Tied advisers would have to disclose contractual ties to the FSA. This is
an ongoing cost. However, this is not expected to be a significant incremental
cost, because tied advisers already have to adhere to a similar obligation (they
have to disclose their marketing group ties to the FSA).
Cost: Financial advisers would have to be classified as such by the FSA. This
is a one-off cost. Based on how much it costs an IFA to gain authorisation, an
upper bound is about £8,000.81 On the basis that there will be 1,200 FAs, this
implies a one-off cost of at most £10 million However in practice the cost
would be much lower than this, because most FAs will have been IFAs, who
would already have paid these costs.
77 This corresponds to 60,000 tied agents x 60% x £330 * 25%.
78 Source: FSA internal note from Peter Andrews to Kevin James, 7 August 1998.
79 This corresponds to 60,000 tied agents x 60% x £2,000 * 25%.
80 This corresponds to 4,000 IFA firms x 5% x £10,000. The cost of notifying customers was provided by
the Association of Independent Financial Advisers.
81 We estimate that it costs an IFA roughly £8,000 to get FSA authorisation. This is made up of £3,000
paid to the FSA as fees, £150 of exam entry costs, and 6 weeks of exam study time which costs the IFA
£4,500. (A new IFA usually has an annual turnover of £35,000. 6 weeks of an IFAÕs time is roughly an
eighth of a year and hence costs £35,000/ 8 = £4,500. Source: www.cii.co.uk) Hence total costs are
£3,000 + £ 150 + £4500 = £7,650.
87
London Economics
July 2000
Cost: FAs will have to adhere to FSA regulations. This is an ongoing cost.
However, since most FAs are likely to have been IFAs previously this is not
expected to be a significant incremental cost.
Benefit: Tied advisers no longer have to adhere to the reciprocity obligation.
Efficiency of competition
Costs: Because our analysis of the market outcome suggests that the number
of IFAs would not be significantly reduced, there are no costs in terms of
reduced competition in the IFA channel, or in terms of reduced competitive
pressure from IFAs on the tied channel.
Benefit: Redefining polarisation puts multiple product providers who

distribute through a FA in direct competition with each other. This could
improve the efficiency of competition, since at present there is little
competition through tied channels.
Benefit: Redefining polarisation lowers barriers to entry for providers who

do not plan to provide all products. As a result, the efficiency of competition
might improve.
Benefit: Redefining polarisation allows providers who are already in the

market, but do not provide the whole spectrum of products, to access more
distribution channels. This could improve the efficiency of competition.
Benefit: Redefining polarisation allows providers who provide many, but not
all, products to plug gaps in their product range without having to incur the
costs of outsourcing activities or re-insurance. This could improve the
efficiency of competition.
Benefit: Redefining polarisation allows providers who are already in the

market and selling a particular product to tie-up with a competitor and sell
the competitorÕs product. As a result this might heighten the efficiency of
competition:
• for the competitorÕs product; and for
• own brand product.
Effect on quantity
Benefit: the quantity sold through tied channels should rise, since the
providers have more products to sell and can provide a suitable product for
more needs.
88
London Economics
July 2000
Effect on quality
Potential cost: If IFAs choose to become FAs, the number of independent

advisers is likely to fall. As explained in the market outcomes section,
however, our analysis suggests this effect will be small.
Cost: When polarisation is redefined, customers may be sold a product by a

tied adviser with a higher commission under the mistaken belief the adviser
is independent. Our analysis, however, suggests that this is unlikely to occur,
because advisers will compete on choice but they will:
• have to adhere to the Òas good as any other product within the
product rangeÓ obligation; and
• product providers who allow a company to use their brand to sell

their own products82 are likely to end up devaluing both brands.
Since this is clearly sub-optimal for the product provider they will
seek another distribution channel before such a market outcome is
realised.
Benefit: The quality of the products produced by providers who might sell
their own products, alongside a competitorÕs, may improve.
Benefit: The quality of the products sold by providers, who are plugging
gaps in their own portfolio, might improve.
Benefit: The quality of products sold by providers who have one or a few
products might improve. This is because now they will be competing to seek
the distribution channel providerÕs favour.
Effect on variety
Benefit: As noted in the Efficiency of competition section, the main effect of

redefining polarisation will be to improve the diversity of products that are
offered by a given tied adviser. This benefit comes in three forms:
• new entrants bring new innovative products to the market;
• tied advisers can offer a wider product range in a given product

category; and
• tied advisers can offer products that they were hitherto unable to
offer.
82 This is sometimes called the Òhalo effectÓ.
89
London Economics
July 2000
9.2 Scenario 2: The DGFT proposal
In this scenario collective investment schemes (CIS) will be exempted from

the polarisation rules. As a result, product advisers will be able to tie with
multiple providers of these products.
In this scenario, the CIS market will be non-polarised. As a result, advisers:
• can tie with multiple providers CIS products; and
• do not have reciprocity obligations with regard to CIS products.
The FSA estimates that this cost will amount to approximately £40,000 as
above.83
Cost: Removing CIS products from polarisation creates regulatory differences

between products that are otherwise fairly close substitutes e.g. unit linked
bonds (a polarised product) and unit trusts (a non-polarised product). This
could result in the financial services market fragmenting into smaller markets
in which the efficiency of competition is less. However, our analysis indicates
that the number of providers remaining in each market segment is so large
that this is unlikely to impose a significant incremental cost.
Cost: Unit trust providers may try to incentivise IFAs (who are independent
for life products) to tie to their unit trust products, and recoup the cost of
these incentives by charging customers higher prices. As a result charges
might rise for unit trust products. If the increased charges are equivalent to
even 1% of premia, the increase in ongoing costs would be £56 million.84
Benefit: The exemption lowers barriers to entry for providers who plan to
enter the CIS market but do not plan to provide other products. As a result
efficiency of competition might improve.
84 At present sales of unit trust products are £14 billion [Source: UK Fund Industry Review Directory
1997 total gross retail sales]. 1% of £14 billion x 40% = £56 million. The increase in premia is smaller
than the difference between an upper quartile fund and an average fund.
90
London Economics
July 2000
Benefit: The exemption allows providers who are already in the CIS market,
but do not provide the whole spectrum of products, to access more
distribution channels for their CIS products.
Benefit: The scenario allows existing CIS providers to tie with competitors
and sell their products. This might heighten the efficiency of competition
both:
• for the competitorÕs product; and
• for the own brand product.
Effect on quantity
Benefit: If competition is heightened, and this lowers product prices, then the
quantity of CIS products sold should rise.
Benefit: The quantity sold through tied channels should rise as well, since:
• distribution channel providers have more types of CIS to sell; and
• CIS providers have more channels via which to distribute their

products.
Effect on quality
Cost: Consumer research shows that at the moment approximately 80% of

consumers using IFAs understand clearly how they differ from tied advisers.
However, this scenario will allow advisers who are independent for life
products to be tied for CIS products and still call themselves IFAs. This
blurring of the distinction between independent and tied advice might create
two sets of costs:
1. It might become possible for IFAs to sell products with high commission
(from providers with whom they have contractual links) to consumers
who believe that they are getting independent advice. In particular,
consumers may go to an IFA for general advice and not appreciate the
fact that swapping between products means that they are no longer
receiving independent advice based on surveying the whole market.
2. Products which are fairly substitutable might fall on either side of the
polarisation divide. E.g. unit trusts will be non-polarised and unit linked
bonds will be polarised. Advisers who are IFAs for unit linked bonds and
tied for unit trusts may be incentivised by providers to recommend that
the customer buys the unit trust. As a result, consumers might end up
91
London Economics
July 2000
buying the wrong product. We estimate the ongoing cost at £48 million
per annum.85
Cost: Consumers using a tied adviser who offers competing CIS products
might buy the wrong CIS product, but this appears unlikely, because:
• distribution channel providers who try to use the brand of a

competitor (who is distributing through them) to sell their own
products86 are likely to end up devaluing the competitorÕs brand. This
is clearly sub-optimal for the competitor, who will seek another
distribution channel rather than face such a market outcome;
• it is relatively easy for some customers to compare across CIS

products and switch between (and out of) them.
Benefit: The quality of the CIS products produced by the distribution channel
providers who might sell their own CIS products alongside a competitorÕs
might improve with the exemption.
Benefit: The quality of the CIS products sold by the distribution channel
providers who are plugging gaps in their own portfolio, might improve.
Effect of variety
Benefit: New entrants can bring new CIS products to the market.
Benefit: Tied advisers can offer a wider CIS product range in a given product
category.
Benefit: Tied advisers can offer CIS products that they were hitherto unable
to offer.
9.3 Scenario 3: Minimum standards exemption
In this scenario, the simplified products market will be non-polarised and

advisers:
• can tie with multiple providers for the same simplified product;
• can tie with different providers for different simplified products; and
85 Annually £40 billion of life and pensions business is written [FSA Returns] IFAs account for roughly
60% of these sales [Association of British Insurers]. Of this £24 billion (for the purposes of an upper
bound) we assume that one tenth are mis-sold CIS products. We assume those who are mis-sold bear
additional costs of 2% of premia through higher charges, resulting in a total cost of £48 million per
year. 2% of premia reflects the differences in RIY between pension and unit trusts.
86 This is sometimes called the Òhalo effectÓ.
92
London Economics
July 2000
• do not have reciprocity obligations for simplified products.
The CBA for this scenario is in some ways quite similar to that for Scenario 2.
The key differences are that:
• efficiency of competition is not hampered in this scenario to the

extent that it is in the previous scenario; and
• the potential for providers to confuse consumers and sell higher price
or lower quality products is lower in this scenario as compared to the
previous scenario.
As a result, the net effect is likely to be positive. However it is also likely to be

very small. The caveat to this is whether this allows a much wider
distribution of stakeholder pensions as providers can benefit from exploiting
economies of scale of the large providers.
Cost: The FSA would have to rewrite the polarisation rules. This is a one-off
cost. The FSA estimates that this cost will amount to approximately £40,000 as
above.87
Benefits: Reciprocity obligations would be abolished in a simplified products

market encouraging providers to offer multiple products.
Cost: Regulation is likely to impose differences between products that are

otherwise fairly close substitutes e.g. unit trust ISA and unit trust CAT
marked ISAs. It might be argued that this might result in the financial
services market fragmenting into smaller markets in which the efficiency of
competition is less. However, at present simplified products such as CAT
marked ISAs are already a distinct market segment. Hence introducing this
scenario is unlikely to have a significant adverse effect on efficiency of
competition.
Benefit: The exemption lowers barriers to entry for providers who plan to
enter the simplified products market but do not plan to provide all products.
As a result, the efficiency of competition might improve.
93
London Economics
July 2000
Benefit: The exemption allows providers who are already in the simplified
product market but do not provide the whole spectrum of products to access
more distribution channels for simple products. Two types of providers fall
into this category:
• simplified product providers with few products or a single product;

and
• simplified product providers with many, but not all, products.
Benefit: The exemption allows simplified product providers who are already
in the market and selling a particular simplified product to tie-up with a
competitor and sell the competitorÕs product. As a result this might heighten
the efficiency of competition:
• for the competitorÕs product; and for
• the own brand product.
Effect on quantity
• the distribution channel providers have more simplified products to

sell; and
• the single simplified product providers have more channels via

which to distribute their products.
Effect on quality
Cost: Research shows that at the moment consumers using IFAs largely
understand the difference between IFAs and tied advisers. However, this
scenario will allow advisers who are independent for complex products but
tied for simplified products to call themselves IFAs. This blurring of
distinction between independent and tied advice might create confusion. In
particular, it might become possible for advisers to sell products with high
commission (from providers with whom they have contractual links) to
consumers who believe that they are getting independent advice. In
particular, consumers may go to an IFA for general advice and not appreciate
the fact that swapping between products means that they are no longing
receiving independent advice based on surveying the whole market. While
this market outcome is possible, it is not obvious that there are incremental
costs arising from it. This is because (as explained in the market outcomes
section) it is not possible to charge very higher charges on simple products.
Benefit: The quality of the simplified products produced by the distribution

channel providers who might sell their own simple products alongside a
competitorÕs might improve.
94
London Economics
July 2000
Benefit: The quality of the simplified products sold by the distribution

channel providers who are plugging gaps in their own portfolio might
improve.
Effect on variety
Benefit: New entrants will bring new simplified products to the market.
Benefit: Tied advisers will be able to offer a wider simplified product range
in a given product category.
Benefit: Tied advisers will be able to offer simplified products that they were
hitherto unable to offer.
9 . 4 Scenario 4: Retain the essentials of

polarisation
The benefits arising from this scenario are similar in nature to those from
Scenario 1. However, because tied advisers are allowed to tie to only one
producer per product, the benefits are smaller in magnitude than in Scenario
1.
The FSA estimates that this cost will be about £40,000.88
Cost: The FSA would have to monitor gap-filling to check there is only one
product in each product category. There is a one-off costs in defining the
product categories and one-going cost in monitoring providers.
Cost: Tied advisers would have to disclose contractual ties to the FSA. This is
an ongoing cost. However, it is not expected to be a significant incremental
cost because tied advisers already have to adhere to a similar obligation (they
have to disclose marketing group ties to the FSA).
Benefit: Tied advisers would no longer have to adhere to the reciprocity

obligation.
95
London Economics
July 2000
Benefit: Permitting gap-filling lowers barriers to entry for providers who do

not plan to provide all products. As a result efficiency of competition might
improve.
Benefit: Gap-filling also allows providers who are already in the market but
do not provide the whole spectrum of products to access more distribution
channels. As a result efficiency of competition might improve.
Benefit: Gap-filling allows providers who provide many, but not all,
products to plug gaps in their product portfolio without having to incur the
additional costs. As a result efficiency of competition might improve.
Benefit: Gap-filling allows providers who are already in the market, are
already selling a particular product, but are unhappy with its performance, to
tie-up with a competitor and sell the competitorÕs product instead (without
incurring the additional costs of re-insurance or outsourcing activities). As a
result less profitable products are likely to be taken off the market and
replaced by more profitable products. To the extent that this leads to the
removal of inefficient products from the market, efficiency of competition is
improved.
Effect on quantity
• the gap filling distribution channel providers have more products to

sell;
• the distribution channel providers who have replaced their own

ÒinefficientÓ product with a competitorÕs product has a better product
to sell; and
• the single product providers have more channels via which to

distribute their products.
Effect on quality
Cost: If IFAs choose to become tied advisers, the number of independent

advice providers is likely to fall. Our analysis however suggests that this in
an unlikely market outcome.
Benefit: The quality of the products produced by the distribution channel

providers who might replace some of their own products with competitorsÕ
products, might improve.
Benefit: The quality of the products sold by the distribution channel

providers, who are plugging gaps in their own portfolio, might improve.
96
London Economics
July 2000
Benefit: The quality of products sold by providers who have one or a few
products might improve. This is because these providers will now be
competing to seek the distribution channel providerÕs favour.
Effect on variety
Benefit: New entrants will bring new products to the market.
Benefit: Tied advisers will be able to offer products that they were hitherto
unable to offer.
97
London Economics
July 2000

Polar

Uploaded by

Copyright:

Available Formats

Polar

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Polar

Uploaded by

Copyright:

Available Formats

Polarisation and

A review for the

2 Introduction and outline 4

3 The current regulatory environment 6

4 The market for polarised packaged products 15

5 The economic effects of polarisation 25

6 Technology and regulation 49

7 Options for change 56

8.1 CBA summary tables 84

9 Appendix: Detailed CBA 86

Table 1: The changing reduction in yield (RIY) on life and

Figure 1: Distribution of life and pensions new business (APE),

Figure 10: Extent of understanding of polarisation status by

This report provides a review of the polarisation rules and an analysis of

• Redefinition of polarisation for all products. In this scenario, firms

• Retaining the essentials of polarisation. In this scenario, providers

In addition, we considered two variants, namely the separation of product

The key analytical consideration is therefore whether polarisation can be

IFAsÕ different contractual relationship with their client. We conclude

The results of our cost-benefit analysis of the scenarios are as follows:

p Redefinition of polarisation for all products. Although such a change

p The DGFT recommendation. This scenario may reduce the effectiveness

p Minimum standards exemption. This scenario is less detrimental than

p Retaining the essentials of polarisation. This scenario offers some gains

In our view, three of the core scenarios (redefinition, minimum standards

2 Introduction and outline

The regime whereby advisory roles with respect to packaged investment

p The redefinition of polarisation for all products, so that firms could

In addition, we considered two variants, namely the separation of product

In addition to the submissions made to the FSA, LE has drawn on a number

end up using different distribution channels. The survey was

The rest of this report is structured as follows:

3 The current regulatory environment

The current regulations governing the provision of financial advice are

• the market and regulatory context at the time polarisation was

3.1 The introduction of polarisation

• agents of insurance companies, who worked solely for that company;

• insurance brokers; and

• firms or companies paid through commission for introducing

The Act put into place a comprehensive system of regulation, covering

2 With particular reference to CP45A.

• less investor confusion about advisersÕ status;

The arguments against polarisation were that it would prevent, restrict or

• costly disruption of commercial arrangements, requiring large

• a significant reduction in the numbers of people who were able to

3.2 Description of the polarisation regime

The polarisation rules require that:

Ò1. A firm which advises a private customer on packaged products must

a) be a product company or its marketing group associate; or

2. A firm which is a product company or its marketing group associate must

3. A firm which acts as an independent intermediary in advising a private

Polarisation only applies to advice on packaged investment products. These

• personal pensions; and

• collective investment schemes, investment trust savings schemes,

3.2.1 Independent advice

An IFA can be the appointed representative of an IFA network. In this case,

3.2.2 Tied advice

• appointed representatives of a product provider; or

• company representatives employed by the providers.