Stakeholder Pensions: The Hidden Charges

Home page

Ten stakeholder myths

1. The cap con

On the website of the DWP it says:

"A stakeholder pension scheme cannot charge more than 1% a year on the value of each member's funds." ^x

The booklet of the DWP Stakeholder pensions - Your guide states:

"As well as the one per cent, the law allows pension providers to recover costs and charges they have to pay for certain other things. For example, when they have to pay any stamp duty or other charges for buying and selling investments for your fund, or for particular circumstances such as the costs of sharing a pension when a couple divorce. These expenses are found in other pension schemes, not just stakeholder pensions. ^{( page 6 )}

This text appeared in the booklet for the first time in April 2001 when stakeholder pensions were introduced on the market. This is saying that providers can make charges additional to the one per cent to recover certain costs, which conflicts for example with:

"The most you can be charged each year by the organisation running your scheme is 1% of the amount in your fund."; "The maximum annual management charge will be 1% of the pension fund each year."; "There’s just a 1% maximum yearly management charge."; "There's a maximum annual charge of 1 per cent of the value of your fund,"; "There is only one charge payable on a stakeholder pension known as an Annual Management Charge."Stakeholder pensions cannot charge more than 1pc a year"; " Schemes may only charge a maximum of 1% per year of funds accrued. . . . stakeholder fund managers can only charge a maximum annual management fee of 1% of the entire fund. They cannot charge you for anything else. "; "Our only charge is a 1% annual fee, falling to 0.7%. "

Dealing costs are discussed in an article in The Sunday Times. ^x This estimates dealing costs to be up to 1.3 per cent of capital per annum, quoting the FSA. Dealing costs arise from portfolio turnover. The 1 per cent cap on charges has run into the problem of excessive portfolio turnover in the financial services industry.

The BBC Radio 4 Money Box ^x programme investigated "the true cost of stakeholder pensions". In the broadcast on 28th April 2001, the Legal and General representative said that it is "complying with disclosure requirements". They need changing. If only one charge has to be disclosed then companies are allowed to say that there is only one charge, when they mean that there are other charges, but we are not telling you about them. To make companies disclose the existence of the additional charges will require new regulations.

"Charges for buying and selling investments for your fund", that is dealing charges, were not discussed in the stakeholder consultations. They suddenly appeared as a charge outside the 1 per cent cap in the (draft) Regulations - 14(5)(c). Now that stakeholder pensions have been introduced, they have appeared for the first time in the Guidelines. For actively managed funds, it is estimated in Section 5 below that further costs and charges will average an additional 1.5 per cent. So that total costs and charges are 2.5 per cent per annum.

Various questions need to be asked about the stakeholder cap, limit or maximum. What do the Regulations say? How does this compare with the cost of occupational pension schemes? Will stakeholder schemes have accounts available on the internet? Such accounts need to specify what are the charges, including dealing costs, to show that they do not exceed the one per cent cap - or at least which are included in the cap.

2. Four levels of charges

The FSA's Conduct of Business (COB) rules distinguish between "charges" and "expenses". Charges are explicit payments made to the provider. In the case of personal pensions the main ones seem to be:

the main management charge which is a percentage of capital per annum

ditto for an initial period - probably for the broker's commission

fixed amount per month

initial percentage of capital

set-up fee of a fixed amount

exit fee as percentage of capital if a policy is cashed in early

"Examples of expenses are:

(a) registration fees;

(b) safe custody fees

(c) trustees' fees;

(e) audit fees;

(f) regulators fees and subscriptions;

(g) costs of investment management but excluding dealing costs of the underlying portfolio, and costs associated with routine management and servicing of existing property investments;

(h) bid/offer spread in the pricing of units

(3) The spread in (h) should be on a basis that fairly represents the expected levy of such spread in a firm92s experience of normal trading conditions.

(4) The expenses should include allowance for any value added tax which is not recoverable." (COB 6.6.67)

Expense charges in general are discussed in the October 1999 consultation paper of the FSA Comparative Information for Financial Services which states:

"Not all charges are apparent to the consumer. Some of them are clearly revealed, others are revealed but in a fashion that does not allow the total price of the product to be assessed easily. But some, most notably the dealing costs of managing the portfolio are hidden."

This indicates that there are four level of charges of increasing stealth: 1) the main management charge, 2) fees in the small print in accounts, 3) dealing costs, 4) other hidden costs. In the case of unit and investment trusts 1) and 2) combine into the total expense ratio (TER). 1) and 2) are explicit or disclosed charges, 3) and 4) are implicit or hidden charges. The small print in the accounts 2) is the COB charges and expenses, except for the main management charge 1).

Life assurance policies do not have accounts like unit and investment trusts. In this case the small print is in the policy documents. The expenses are not given but are combined into the reduction in yield (RIY) indicator.

There are discussions in the press about 1) and 2) in comparison to personal pensions. But 3) and 4) are neglected. This is the purpose of this website. We need first to know what are the various charges,before we can say they are being capped. Otherwise there is not proper transparency.

Hidden charges divide into those that come out of capital and those that come out of investment income. Dealing charges come out of capital. They are concealed in the selling price of shares to fund managers. Because unit trusts have accounts and pay dividends, it is difficult for them to take hidden charges out of income. This is not the case for insurance funds which do not have accounts or distribute dividends.

Since the Labour government came to power in 1997, hitherto hidden charges of the financial services industry have been coming to light. In his 1999 budget speech the Chancellor, Gordon Brown, reported that he had asked the FSA to prepare league tables of the costs and charges of products of the financial services industry. This led to the February 2000 occasional paper of the FSA The Price of Retail Investing in the UK, which is a study of the cost of investing in the retail sector, that is products sold by the industry to individuals.

The maximum 1% charge is the provider's management charge. If stakeholder schemes have trustees, they will apparently be remunerated by providers out of the maximum 1% charge, rather than by the scheme. This gives an incentive to trustees to act on behalf of the providers rather than the members of schemes.

A 1996 report by Timberlake & Company Comparative Total Expense Ratios ("TERs") For International Funds from a survey of unit and investment trusts states:

"Whilst investors might reasonably assume that a fund charging 1% per annum is less expensive than one charging 1.5%, this is often not the case. Publication of management fees as an indicator of the "cost" of fund management is totally misleading. Management fees should not be published on their own, but total expense ratios should always be quoted."

So should dealing costs. The February 2000 occasional paper of the FSA states:

"Furthmore, disclosed charges alone provide a misleading impression of the price of investing as they do not include dealing costs (the cost of trading in a fund's portfolio). . . Consequently, retail investors now lack the information and knowledge they need to make informed choices between the funds they must choose between."

The green paper A New Contract for Welfare: Partnership in Pensions (1998), promised:

"A simple charging structure which members can understand and which allows ready comparisons between schemes."

Without knowing dealing costs it is not possible to compare schemes.

The green paper promised that stakeholder pensions would conform to CAT standards. "For stakeholder pension schemes, compliance with the standards will be a requirement for the scheme to be designated as a stakeholder pension scheme". The Treasury asks: " Will there be CAT standards for stakeholders pensions or mortgages"? Stakeholder pensions are said to "fall in line" with CAT standards. But there are differences between the stakeholder standards and CAT standards for ISAs.

The units of stakeholder schemes have to have single pricing under regulation 8(4). But there is no mention of a dilution levy. It seems that members transferring between schemes may incur a dilution levy when selling units in one scheme and purchasing them in another, even though it is claimed that such transfers incur no charges. It is not realistic to have absolutely no charges for transfers. Someone has to pay. No charges might result in excessive switching between schemes, as happens under the Chilean system in Latin America.

3. Dealing costs and charges

3.1 "A simple and transparent charging structure"

Various reasons are being presented for not mentioning dealing charges, when saying there is a cap on charges.

1: Dealing costs are common knowledge.

This is untrue as discussed in the article of The Sunday Times Unit trusts charge billions in secret fees (1st October 2000). Dealing costs were not part of the 1995 PIA Disclosure Regime of charges. They are referred to for example by the FSA as a "hidden", "implicit" or "undisclosed" charge.

2: Dealing costs are inevitable.

This is all the more reason for including them in the cap. Charges divide into those that are inevitable, and those that are optional which are for life insurance and for financial advice.

3: The 1 per cent limit is on "charges", which does not include "costs".

Charges are a cost on whoever is being charged. But not all costs are the result of charges, so that costs are more general than charges. The bid/offer spread results in a cost for dealing which is not a charge, whereas stamp duty and stockbrokers' commissions are charges. The new DWP (DSS) guidelines refer to "costs and charges they have to pay for certain other things". But surely providers do not have to pay for the bid/offer spread.

If we distinguish between costs and charges, investors then need to know what are the total costs. This depends on the level of discussion. On the first level we can say that everything is a charge, either explicit or implicit, as in Section 2 above. On the next level we can distinguish between costs and charges. There need to be agreed definitions. Whatever these are, it is not fair to talk only about charges, and then say there are further costs after someone has bought a policy.

People are being told that stakeholder schemes have a simple and transparent charging structure, which is far from the case. There is a similar situation with unit trusts. The Sunday Times in its article states:

"'We were reassured because the charging structure appeared to be simple and transparent.'. Unit trusts are, however, not competitive, simple and transparent. . . the unit trust sector has been quietly lifting its charges to mop up the tax breaks the chancellor intended for investors."

Dealing charges are the reason given by The Sunday Times why the charges of unit trusts are not "simple and transparent". They are hidden, since they are not specified in the accounts of unit trusts. They are not part of the 1995 PIA Disclosure Regime of charges. Investors find they have been paying dealing charges which they had not previously realised existed. Probably in years to come further hidden costs will be revealed, in the same way that the existence of dealing charges has been revealed.

There have been extensive consultations about the 1 per cap on charges, but dealing charges in 14 (5) (c) were introduced into the draft Regulations without having been part of the stakeholder consultations. Consultation Brief 1 of the DWP (DSS) is concerned with charges but, whilst stressing the need for a "simple and transparent charging structure", does not mention dealing charges.

This is a serious omission, since average dealing charges for unit trusts are estimated for example in the Financial Times to be 1.2 % of capital per annum. Stakeholder schemes may have similarly high dealing charges, especially if invested in unit trusts. Such high charges result from excessive portfolio turnover.

3.2 Excessive portfolio turnover

"Investors as a group cannot outperform the market because they are the market." According to the OFT, the main merit of index-tracking is that it reduces fund turnover. It is not important for the choice of shares to exactly correspond to a particular index, so that "passively managed funds merely needs to be representative of the market as a whole. A buy and hold strategy will suffice".

An article in U.S. News finds that turnover should not be more than 20% per annum, and funds with higher turnover have a worse investment performance. It distinguishes between investing and playing the market - that is gambling: "When you have portfolio turnover of more than 100 percent, it makes little difference what you know about a company, because you'll own it for less than a year".

The amount of information about the portfolio turnover of mutual funds in the United States is massive. An internet search on "mutual fund turnover", produces thousands of references (including the turnover of fund managers). Mutual funds in the United States have to publish such information for investors - unlike unit trusts in this country. Fund turnover is published by the CSO in Financial Statistics. For 1993 and 1998 we find:

Table 1

Fund turnover

	unit trusts	investment trusts	insurance companies*	pension funds
UK ordinary shares £ bn
end year 1993	33	14	166	243
purchase + sales	38	9	60	93
turnover % in 1993	58	32	18	19
overseas shares £ bn
end year 1993	22	17	46	82
purchase + sales in 1993	52	19	50	85
turnover % in 1993	118	54	54	52
UK ordinary shares £ bn
end year 1998	93	23	305	335
purchase + sales in 1998	86	20	118	201
turnover % in 1998	46	43	19	30
overseas shares £bn
end year 1998	51	18	73	109
purchase + sales in 1998	90	20	84	136
turnover % in 1998	88	55	58	62

Note: * "long term funds", turnover= (purchases+sales)/(2 x year end)

The FSA should be ensuring that dealing charges are not excessive. It is proposing to offload responsibility onto consumers with the development of a "forward-looking price indicator" which "incorporates" dealing charges. A league table of funds ranked by portfolio turnover would certainly be interesting, and encourage low turnovers. Apparently the FSA does not like the idea of league tables. It seems that its forthcoming tables for its Comparative Information Scheme, will be ranked by companies in alphabetic order.

Dealing charges can be estimated from portfolio turnover in the above table. The February 2000 occasional paper estimates dealing charge by multiplying turnover by 1.8 % of capital (0.5 % for stamp duty, 0.3 % for broker's commission, 1 % for bid/offer spreads). For unit trusts in 1998, from the table, dealing charges are estimated to be 1.1 = 1.8 x (86+90)/(2x(93+51)) % per annum.

Including investment trusts and insurance companies with unit trusts we obtain 0.67 % for dealing charges for the industry in 1998. But total charges for actively managed funds are about 3% as mentioned above. According to table 19 of the occasional paper, about half of these charges are hidden. The charges for index-tracking funds are less but according to this table nearly half the charges for index-tracking funds are also hidden.

Total stamp duty paid on purchases of shares was £2,488 million for the year 1998-99 from Inland Revenue Statistics. To get total dealing costs we multiply by 1.8 and divide by 0.5 giving £8,957 million.

The retail sector is all products sold by the financial services industry to individuals. The £8,957 includes dealing by individual shareholders and pension funds, which are not part of the dealing of the retail sector. Individual shareholders own 20 per cent of shares. The total hidden charges for the retail sector is estimated in the February 2000 occasional paper of the FSA to be about £15 billion, as discussed in Section 5 below.

Thus the total dealing costs for the retail sector seem to fall well short of the total hidden charges. So that the hidden charges 4) in Section 2 seem to be substantial. They seem to relate to the reinvestment of investment income. They will exist also for stakeholder pensions. Where there are not proper accounts it is not possible to see that income is being reinvested.

The accounts of unit trusts show the main purchases and sales of shares by total value. It would be useful to know the average price per share, or total number of shares bought and sold. Unit holders could then enquire for example: "How is it you bought one million shares of Associated Sprockets Ltd at £1 each and also sold one million at 50 pence."

3.3 Soft commissions

A reason for high dealing charges reported by The Sunday Times in its article is: "Often a stockbroker that is used will turn out to be another division of the same company." So that dealing charges are going to the same organisation.

Another reason it gives is "soft commissions", referred to as "kickbacks" in the US and "backhanders" by The Sunday Times in its article:

"Backhanders, or 'soft commissions', are also allowed by the regulators. 'Soft commission arrangements are common between fund managers and stockbrokers. . . . Basically, it is payment in kind in return for the fund manager giving the broker his business.' " ^x

Soft commissions encourage high turnover. One unit trust accounts describes soft commissions:

"These brokers provide services to the Manager - such as company and economic research and valuation - which may assist the Manager in the effective running of the Trust. The Manager does not make direct payment for these services, and the services are provided only if the Manager transacts an agreed amount of business with the brokers."

The recent Institutional Investment in the United Kingdom: A Review by Paul Myners recommends that soft commissions should be abolished. Soft commissions are said to be mainly for research work performed by the stockbroker for the fund manager. They are reported by the Sunday Times to be common for unit trusts which have, nevertheless, on average badly underperformed markets, inspite of all this research work. The average performance of pension funds according to statistics of the WM Company, has on the contrary been as good, or even slightly better than, share indices.

3.4 How should high turnover be discouraged?

To discourage churning, that is an excessively high turnover of shares, the sale of shares by a fund should arguably be prohibited, if it is not having to realise its capital, or perhaps sales should be limited to say ten per cent of its portfolio per annum. When such shares are sold investors in general do not benefit, but they have to pay the dealing costs. Conversely, if a fund is declining, there should be a limit on purchases.

Such a rule against the sale of shares would focus attention on which shares to buy. The Myners Review refers to soft-commissions being paid by the sell-side. If sales were largely prohibited this would discourage the payment of soft commissions, which should not in any case be allowed.

There could be a special category of actively managed fund for which turnover is restricted by law. Some investors may prefer funds with a high turnover. But this is offloading responsibility onto investors. If high turnover is a disadvantage for investors, then it should be restricted by law.

In conclusion, churning is a major problem, and needs to be discouraged. Four ways of doing this are: 1) include some or all dealing charges in the cap on charges, 2) better disclosure of dealing charges, 3) abolish soft commissions, 4) restrict turnover by law.

4. Who is making the charges?

4.1 Schemes or providers

Financial service companies are often called "providers". But there seems to be no mention of either "financial service companies" or "providers" in either the stakeholder Regulations or Welfare Reform and Pensions Act. Contributions are paid to "schemes" rather than to "providers". An employer designates a "scheme" rather than a "provider". Charges are imposed by "schemes". For example regulation 3 (a) refers to "the amount of charges which may be made by a stakeholder pension scheme". Therefore:

"the law allows pension providers to recover costs and charges they have to pay for certain other things. For example, when they have to pay stamp duty . . "

in the latest DWP (DSS) Guidelines, is not very exact, since the law does not mention "providers". "Recover" means to make charges to pay for. Stamp duty is taken from your savings and paid over to the government by stockbrokers, rather than by providers.

4.2 Policies or contracts, managers or trustees

The one per cent limit seems to apply only to the explicit charges specified on contracts. Dealing charges are not specified on contracts. They are implicit. They cannot be known at the time when a pension is sold, because they are in the future. The August 2000 consultation paper of the FSA The regulation of stakeholder pensions gives the following definition of "charges":

"For stakeholder pension schemes, charges are all explicit charges and expenses for the underlying policy or contract including any charges levied by the manager or trustees of the stakeholder scheme." (6.6.20)

"There is a limit on the management costs which stakeholder providers can charge." (Annex E 2)

The stakeholder charges or "reductions" listed in the Regulations, do not state who they are made by, incurred by, or paid to. The cap in the Regulations 14(2)-(4) is on "deductions". For investments in unit trusts, oeics, and investment trusts via IPAs, are the various fees paid to, or levied by: the IPA manager, auditors, custodians, registrars, regulators, trustees; included in the one per cent limit? These are charges in the small print, in the second level of charges 2), in Section 2 above.

A reduction in one stakeholder key features document, but not mentioned in the Regulations, is the market value adjustment which can be applied in some circumstances: "Examples of such circumstances would be a prolonged or deep fall in stock market values". Exactly how large this fall has to be is not specified.

4.3 Plans or funds

An article on stakeholder pensions in the Financial Times (March 3-4) states:

"To qualify for stakeholder status, a pension plan must . . . have annual charges of 1 per cent a year or less. (Financial advice can be charged for separately, and the funds in which you invest may incur other costs such as dealing charges.)"

This article and further articles in the Financial Times, seems to be the only time that the existence of dealing charges outside the stakeholder cap was mentioned in the press before it was put in the Guidelines.

The funds could be part of the life fund of the provider. Costs incurred by the funds are then the same as costs incurred by the provider. Regulation 10(3) refers to: "The manager of the scheme, and any person investing funds held for the purposes of the scheme". This suggests that there is a separate fund manager, which indeed there should be. He can then be dismissed for poor performance. Your funds are then being managed by fund managers rather than by providers.

But there may not be separate fund managers. The TUC stakeholder scheme has three funds managed by the Prudential (and two by Standard Life). The Prudential is the "administrator" of the scheme. The TUC has a five year contract with the Prudential to administer its scheme.

A member of the Prudential scheme may find investment performance is unsatisfactory and move to another scheme, but by then he or she will have lost some capital. The TUC scheme has trustees. They will not be able to change the managers of the funds, but can change the administrator after the five years. What are the duties of the administrator? Should the administrator not be independent of fund management?

Whoever is looking after your funds should be acting on your behalf, rather than on their own behalf or on behalf of shareholders. Stakeholder schemes will nearly always have providers. They are the people most responsible for looking after your funds. There should always be properly appointed trustees. But trustees alone are not an adequate safeguard. They will not be able to change providers, at least without a heavy penalty.

In the financial services industry, there are many closed funds, which are no longer being promoted to the public. Some stakeholder schemes and their funds may become closed in time, which reduces the incentive for good investment performance.

The accounts of large occupational pension schemes, such as the universities' USS scheme have a separation between administration and investment costs. The March 2000 accounts give assets of over £21 billion. The "administration costs" are £6.5 million and "investment management expenses" £14.3 million. The USS does not of course have to pay the 1 per cent maximum charge of stakeholder pensions. One per cent of £21 billion is £210 million per annum. How much would the universities' former insured FSSU scheme be paying in expenses, if it still existed?

The maximum 1 per cent charge is described as a "management charge", for "routine administration" in: "The only charge we make for the routine administration of your plan is an annual management charge. The maximum charge is about 1% per annum but this reduces as your fund grows.". This suggests that there is a separate charge for fund management. There is the possibility that the salaries of fund managers in some schemes will be paid out of dealing costs under regulation 14 (5) (c).

The Building and Civil Engineering (B&CE) scheme states that there are no charges until at least 2006. "It won’t cost you a penny in charges until at least 2006." - for workers in the construction industry who are members of its employee benefit scheme, otherwise charges are 0.65 per cent per annum. Funds are managed by Barclays Global Investors. Does this 0.65 per cent include charges for fund management? The Financial Times reports that sales of stakeholder pensions have been disappointing (14th August 2001, front page), and that in the first three months the B&CE scheme has "60 per cent of the market, dwarfing the share of the established names".

4.4 Custodians

If the investments of a fund have a high turnover, certificates are continually being transferred in and out of custody, which greatly increases the custody fees. The DSS said at one time that custody fees attributable to dealing could count as dealing costs - in which case they fall outside the 1 per cent cap. It may be difficult to determine which part of the fees are associated with dealing. The DWP (DSS) has now changed its mind and thinks that all custody fees should fall within the cap.

The ABI had discussions with the DSS on this topic. The ABI states that "in their opinion all custodian fees should fall within the 1% charge", but "ultimately interpretation of the Regulations is a legal matter". This probably means that, because there is official uncertainty, custody fees will not be included in the cap - at least until there is a legal challenge.

Who are the custodians? In the case of private investors investing in foreign shares, the shares are often held in nominee accounts, which stockbrokers describe as "custody". The charge is typically 0.4-0.5% of capital per annum. Do stockbrokers act as custodians in this way also for institutional investors?

4.5 Stockbrokers

The Myners Review states that stockbrokers commissions are not properly disclosed by fund managers to their clients: "although they are disclosed, this is done in a way which is far from transparent." He recommends that these commissions should be included as part of the management charge made by fund managers to their clients:

"Clients' interests would be better served if they required fund managers to absorb the cost of any commissions paid, treating these commissions as a cost of the business of fund management, as they surely are."

The Financial Times has had a series of articles on the Myners Review. In an article on 31st May, a pension fund consultant Michael Roberts, states:

"the system, where fund managers negotiate the commission rate with brokers and then get the pension fund to pay the bill, is flawed. It lacks transparency, not least because pension funds are not party to the negotiations over the commission rate."

How many members of stakeholder pension schemes will be party to negotiations over commission rates - or even be aware of what they are? Judging from the stakeholder consultations, which never mentioned these commissions; they are not intended to concern themselves with such matters, even though they are of course paying for them. It follows that stockbrokers' commissions should be included in the 1 per cent cap on charges.

The Treasury states: "Our CAT standards will get rid of the small print and hidden charges that worry people so much". They do not get rid of stockbrokers' commissions as a hidden charge. That is hidden in the sense that investors do not know their magnitude, but their existence is becoming increasingly apparent. The government may have been pursuaded by the Myners Review and articles in the Financial Times to mention dealing charges in its latest stakeholder Guidelines.

The Swedish Shareholders' Association has 142 thousand members. A reason for its success, is that it has its own stockbroker. This is like a mutual stockbroker - which we do not have in this country.

You do not have to be a member of your local market in order to buy fruit and vegetables. But to buy shares on a stock market you have to be a member of the market, that is a stockbroker. It should be possible for investors to buy and sell shares on stock markets, without having to go through stockbrokers.

The present author asked Toby Keynes the Project Coordinator of the United Kingdom Shareholders Association whether stockbrokers are necessary in view of the growth of online trading. He replied that "much of the stockbroker function is becoming unnecessary already". The Myners Review has a section entitled "Brokers' commissions". But are stockbrokers needed?

4.6 The government

The government charges stamp duty, which is part of dealing costs and therefore not included in the stakeholder cap. However stamp duty is included in CAT standards for ISAs. Regulation 38 states:

"The total annual charge specified is the limit on all charges borne by the saver. Where dilution levy or stamp duty on the dealings in units or shares is charged, the annual charge must cover these additional cost(s). The saver must pay nothing extra."

According to footnote 5 on page 7 of the February 2000 occasional paper of the FSA investors are receiving the market rate of return on £600 billion, and the industry the market rate on £300 billion. Total funds invested at end 1998, are estimated to be £900 billion (mainly £700 billion life assurance, £180 billion unit trusts.) The market rate is estimated to be 10% per annum. This implies that the total average charges for the industry is 3.33% of capital per annum, which is £30 billion per annum.

Table 2 of the May 2000 occasional paper of the FSA Saving for retirement: How taxes and charges affect choice is entitled: "Projected funds after £60 a month contribution". The "median charge personal pension" has a reduction in yield (RIY) of 1.63% whereas the "stakeholder pension" has a RIY of 1.06%, making a difference of 0.57%. This suggests that the cap on charges reduces charges by 0.57%. But the RIY 1.63% is only about half of the 3.33% for all charges.

A charge which is a fixed percentage of capital is wrong in principle. Would you let someone manage your business such as a farm, in return for one per cent of the farm each year?

Part 2. "Informed investment decisions"

6. Information

6.1 The cap on charges

The stakeholder helpline run by the Occupational Pensions Advisory Service (OPAS), states that:

"The provider or manager of a stakeholder scheme will normally levy charges to cover their costs and provide a profit. These charges are limited to being no more than 1% of the value of your fund each year."

This will have to be changed in view of the latest DWP (DSS) Guidelines, which say that 1 per cent does not cover all the costs and charges incurred by providers.

The DSS clearly consulted with the Occupational Pensions Regulatory Authority (OPRA) before issuing its latest Guidelines, because it has changed its website accordingly. But OPAS has not. There seems to be something wrong with DWP (DSS) consultation procedures in general. The blue paper Partnership in Pensions stated that responses would be made available to the public. When the present author asked to see them by visiting the DSS or Record Office of the House of Lords, he was told that this was not possible, but he could be sent photocopies of a limited number. Responses to other green papers can be seen in the Record Office. Instead of being able to see the responses, he was sent the addresses of all 518 respondents!

It is possible that not all stakeholder schemes will have providers or managers, since not all occupational money purchase pension schemes have providers or managers. Some stakeholder schemes could in theory be self-administered, as discussed above.

The FSA website also needs to be changed. Dealing charges are not mentioned in its factsheet Stakeholder pensions and decision trees , which states:

"Providers of stakeholder pensions usually charge for managing your money. There is an upper limit on this charge. The limit is 1% of the value of your fund each year. The charge is taken from your fund."

This implies that there are no further charges, beyond 1 per cent. Further websites will need changing for example: "The Government has said that stakeholder pension providers will not be allowed to charge anything other than an annual management fee, capped at 1%."

A phrase such as: "Charges are 0.9 per cent per annum.", on the websites of providers, and for example in key features documents, is misleading unless it is made clear that there are further charges. It therefore needs to be modified, such as by adding: "and there are further charges especially for dealing costs.".

6.2 Disclosure regimes

In a paper on the disclosure regime which the PIA introduced for life and pension products in 1995, the chairman Joe Palmer states:

"A key element of the PIA's approach to investor protection is to ensure that investors are given the relevant information necessary to enable them to take informed investment decisions."

About half of all charges remained hidden under this regime, and it was not even effective in reducing explicit charges. In the Financial Adviser, 2^nd March 2000, page 6, Nic Cicutti manager of the Financial Times website, commenting on the February 2000 occasional paper of the FSA, wrote:

"Any system where disclosed charges amount to barely half of all fund costs is a farce and needs to be reformed urgently."

The disclosed charges in the case of stakeholder pensions are the 1 per cent maximum charge. This implies that there is at least another 1 per cent of undisclosed charges.

The RIY measure of charges was an important feature of the PIA 1995 Disclosure Regime. When first introduced, it was claimed that everything is "all in" - like the stakeholder cap on charges. It turns out that it does not even contain all explicit charges. The October 1999 consultation paper of the FSA states:

"At present the PIA uses the measure 'Reduction in Yield' (RIY), but there are variations in the types of costs that are included in the calculation of RIY. For example, custodians' fees and trustees' fees are currently in unit trust RIY calculations, but not those for life insurance products."

The introduction of the stakeholder cap on charges shows that disclosure of charges together with competition between providers is not effective in reducing charges to acceptable levels. The management charges of unit trusts increased sharply when they were decontrolled in 1979. They were formerly a maximum of about 0.5 per cent per annum. Commission charges increased sharply when the Maximum Commission Agreement was abolished from 1st January 1990. This is shown by the average level of commissions in November 1991, as a percentage of the MCA rate, from Table 11 of the June 1992 OFT publication Savings and Investments Consumer Issues.

Table 2

The increase in commissions resulting from abolishing the Maximum Commission Agreement

Type of investment	Term years	Independents % increase	Representative firms % increase
Regular premium
Enowment policy	10	127	147
	25	126	144
Individual pension	10	123	144
	25	126	140
Single premium
Unit trusts		102	107
Unit-linked bonds		125	137

There are problems with disclosure resulting from the Third Life Assurance Directive of the EU as pointed out by Sir Howard Davies chairman of the FSA, to the Treasury select committee, in its report The mis-selling of personal pensions (12th November 1998):

"I think it is important to say that in the Life area under the third Life Directive, it is not possible to require products to meet certain benchmark standards of disclosure or whatever, but it would be possible to say, 'Here are marks on the bench which are how you ought to disclose your commissions'" (Volume 2,p50)

This directive prohibits the standardisation of documentation. Thus one company can refer to "a cap on charges", and another to "a limit on levies" - which is causing so much confusion with stakeholder pensions. It states:

"Member States shall not adopt provisions requiring the prior approval or systematic notification of general and special policy conditions . . . of forms and other printed documents which an assurance undertaking intends to use in its dealings with policyholders." (EEC92/96, Chapter 3, Article 9, 10th November 1992)

Under the system of self-regulation, disclosure regimes are developed by the industry, which does not want to reveal its charges. A new regime is now being developed by the FSA including its Comparative Information Scheme, which it claims to be an improvement. It will list a considerable number of "indicators" and "sub-indicators" for all the "products" of the industry. Inspite of the many thousands of products and about twenty indicators, the Treasury states that it will enable products to be compared "at a glance".

According to the FSA all product types are an equally good deal and are chosen to suit individual requirements. Product types listed in its October 1999 consultation paper are: personal pensions, investment bonds, unit trust ISAs, savings endowments, mortgage endowments. Its Comparative Information Scheme is intended to help consumers so that:

"once they have decided what sort of product they need, they can make better informed decisions about which particular one might offer them the best deal (1.3). . . . comparative information is intended to help the consumer who has already decided what type of product they are looking for. (5.7)"

All unitised funds selling units to the public and larger pension schemes should have Annual Report and Accounts available on the internet, rather than just indicators, and key features. These should contain information about fund turnover and dealing costs.

To make progress it is necessary to compare product types to see how they perform over time. But the Comparative Information Scheme only provides information about new products rather than existing products. It is designed to promote sales. It will not provide information about past investment peformance, or the benefits being provided by existing products.

There may indeed not be any such benefits because policies remain unclaimed. In the United States:

"More than one-quarter of all life insurance policy benefits go unclaimed and unpaid on death of the insured, due to long dormancy periods and because family members aren't always aware a policy exists."

The industry has recently set up an unclaimed assets register (UAR), which is intended in due course to contain details of all unclaimed life assurance policies.

The UAR will do a search of its register of unclaimed policies in return for £15, if you can show that you have a "legitimate interest", such as being an executor. This is arguably too restrictive. Details of unclaimed insurance policies should be put on the internet, as they are to some extent already in the United States. The names of the shareholders of plcs is not confidential. There is surely no necessity for the names of policyholders to be confidential.

The FSA is in effect saying: "The last disclosure regime was not much good, we are developing a better one". If a product has provided only poor benefits in the past, the industry says: "The last ones were inferior to these new ones, which are much better.". Similarly the government tends to exaggerate the benefits of changes. It is saying that the previous system of self-regulation was unsatisfactory, and the FSA is better, and that stakeholder pensions are "flexible" and "low-cost" in comparison with personal pensions. Stakeholder pensions do not permit "up-front" or "front-end" loading of expense charges which is an advantage, but the new legislation and terminology does not in general solve the underlying problems.

6.3 Accounts

We distinguish between the accounts of stakeholder schemes, and the accounts of the funds in which the schemes are invested. Schemes with trustees have to have audited accounts. But this does not imply that the funds in which the schemes are invested also have audited accounts. There may not be separate accounts for schemes and for funds, because they are enclosed in the same booklet. The accounts of schemes should contain at least:

- the administrative cost of running the scheme,

- and separately the cost of any investment activities,

- the total number of members,

- the total contributions and benefits paid.

All unitised funds in which schemes are invested should have audited accounts, available on request; or these details can be given in the accounts of the scheme, showing:

-	all investments,
-	all investment income,
-	the total value of purchases of shares and the total value of sales,
-	the main purchases and sales giving numbers of shares and at what price,
-	similarly the total value of purchases and sales of other assets,
-	the past investment performance of the fund, for at least the last ten years if possible,
-	all tax paid, such as: stamp duty, insurance tax, any withholding tax on dividends,
-	all charges such as the provider's management charge and stockbrokers commissions,
-	any separate fees not paid for from the expense charges such as: to trustees, auditors,
-	a total expense ratio (TER) calculated from all charges and fees.

7. "The Government looks to the new FSA"

Stakeholder schemes are regulated by OPRA and the FSA. In the March 1998 green paper A new Contract for Welfare there is a "contract" between the government and citizens (page 80):

	Duty of Government		Duty of individual
	Regulate effectively so that people can be confident that pensions and private savings products are secure.		Save for retirement where possible.

The government is not doing its duty to regulate effectively, since it has handed responsibility to the FSA:

"The regulatory process should provide reassurance and protection for customers,... The Government looks to the new FSA to meet this important responsibility for investments such as personal pensions." (page 41, paragraph 32)

When the Labour government came to power in 1997, the Chancellor, Gordon Brown, commissioned a report from the SIB, Reform of the Financial Regulatory System. This is a blueprint for the takeover of the proposed new regulator by the industry, starting with the appointment of the senior executives. The SIB was in charge of the previous self-regulating system. It was delighted to be given extended powers with a change of name to the FSA. It is mainly staffed by the same people as the previous system.

A previous Chancellor also went to the regulators in July 1992 to ask them "what needs to be done to improve regulatory performance", resulting in the report Making the Two Tier System Work, by Sir Andrew Large chairman of the SIB. Judging from these reports, with a system of self-regulation, the regulators should not be responsible for specifying how they can improve their own performance.

The FSA is of course a continuation of the self-regulation of the industry, so that history is repeating itself. It is financed by the industry. It is a quango. The new Food Standards Agency, and regulators such as OFSTED, OFTEL, OFWAT, describe themselves as government departments. Equivalent organisations to the FSA in other countries, such as the SEC in the United States are part of government. The SEC is a federal agency. The Swedish Financial Supervisory Authority has been described as a "trail blazer for the FSA" because it is a single integrated regulator. It is a division under the Ministry of Finance, that is part of government, rather than a quango like the FSA.

OPRA, which regulates occupational pension schemes, is not a goverment department. However there is a big difference between OPRA and the FSA. Employers are largely paying for occupational pensions, whereas insurance companies do not pay for personal pensions.

The Financial Services and Markets Act (2000) has an emphasis on regulating markets, that is on inputs rather than outputs. The FSA regulates the Recognised Investment Exchanges, which includes markets in metals and gas. There should instead be an emphasis on regulating what happens to savings after products have been bought. Instead, as the Treasury select committee stated in its report on the mis-selling of personal pensions:

"Insurance companies hold large amounts of other people's money. We were concerned to learn that the regulation of this aspect of their business is ill-designed, little known and rarely exercised."

Our supervisory authorities of insurance companies have an arms-length approach. They have relied on internal actuaries and now are planning to appoint independent actuaries. The Bundesaufsichtsamt fur das Versicherungswesen (BAV) which regulates insurance companies in Germany states: "As often as possible, the companies are inspected on-site by inspectors who are staff members of the BAV."

The stakeholder Regulations relating to expense charges seem vague. Legislation in this area tends to be vague, containing terms such as "the protection of consumers" in the Financial Services and Markets Act, which are then interpreted by the industry. The protection of consumers is being interpreted by the FSA to be helping consumers choose products to suit their individual requirements, such as its Comparative Information Scheme, which will no doubt be good for sales.

"The protection of consumers" is a slightly confusing expression and should be replaced by "the protection of savings". It should be the duty of the FSA to protect the savings of the public within the institutions which it regulates, not from stock market fluctuations, but from various ills which can befall these savings resulting from for example: insolvency, crime, mis-selling, inefficiency, incompetence, excessive expense charges.

There has been a change from "the protection of investors" in the Financial Services Act 1986 to "the protection of consumers", which has become: "secure an appropriate degree of protection for consumers". The Parliamentary Treasury select committee stated in its report The Regulation of Financial Services in the UK (23rd October 1995):

"The Committee believes that the public interest is best served by a regulatory community . . . which does not rely on the financial community it is supposed to be regulating to set the pace for defining behaviour in key areas of market practice."

The "objectives" for the FSA in the Act are so vague, it will be difficult to determine whether it is achieving them or not. How do you measure "awareness" and "confidence" except by the extent of sales? This vagueness contrasts for example with the objectives of OFTEL: "Our goal is to make sure you receive the best quality, choice and value for money for all your telephone services."

The Financial Instititions and Markets Department of the Swedish Ministry of Finance, states:

"The Department's objective is to create the basis for financial regulation that fosters efficiency in the financial system at the same time that society's need for stability and consumers' interest in being well protected are provided for."

Efficiency and stability are surely better objectives than awarenss and confidence.

The BAV in Germany has a legal obligations to take those actions which it thinks will benefit policyholders. It has always been concerned with protecting the interests of policyholders. "The Office was entrusted with safeguarding the interests of the insured and making sure that contractual liabilities could be met at any time."

Our regulators have been taken over by the industry. This is the well-known phenomenon of "regulatory-capture". They are now concerned with all sorts of issues, other than protecting policyholders, especially those which will help to promote sales.

Sir Howard Davies stated to the Treasury select committee in its report on the mis-selling of personal pensions, mentioned above:

"The whole learning experience of the last few years has taught us I think, that suitability is at the core of the problem which investors face, whether this is a suitable product for you."

The main problem faced by investors is getting a good return on their investments, rather than finding products to suit individual requirements. Savings need to be protected from excessive and hidden charges, which cause the poor average returns discussed in the February 2000 occasional paper of the FSA, mentioned above. They need to be protected from inefficiency. People are being asked to save for retirement in a market which is not properly regulated.

8. The need for a new kind of mutuality

Too much responsibility is being offloaded onto individuals, to cope with expense charges based on inadequate information; to decide whether or not to take out a stakeholder pension, and whether to contract out of Serps/State Second Pension; to choose between: schemes, providers, funds, annuities and even financial advisors.

Decision trees have been developed for stakeholder pensions, which according to the DWP (DSS) and FSA, are to be used as a marketing aid. The first decision which investors need to make is how much time to spend studying the various alternatives. If they make a mistake, this produces feelings of guilt that they did not spend enough time. The FSA states in its October 1999 consultation paper:

"The Government has promised to reform welfare provision and in the future consumers will increasingly need to consider planning their own provision for income protection and retirement income during the course of their working lives. So there will be a growing market, but one that is relatively inexperienced."

People will increasingly be required to make informed investment decisions, which comes close to "picking winners". The government has itself not had a good track record with its informed investment decisions: Ground Nuts, Blue Streak, TSR2, Concorde, atomic energy and so on. An article by Will Hutton and Charlotte Thorne An inequitable pensions policy in the Financial Times (10^th January):

"Britain is coming to rely on an approach to pension provision that expects people to be both extraordinarily forward-thinking and unusually financially adroit when planning their own retirement."

The Treasury states in its report Standards for Retail Financial Products:

"In the modern world people will increasingly have to look after their own financial interests for themselves. No government could realistically hope to meet everyone's expectation for welfare services at a tolerable level of taxation."

In the case of stakeholder pensions your financial interests are looked after by trustees or stakeholder managers. If you are invested in unit trusts, then they are looked after by a unit trust management company. You might sell the unit trust units and buy a life assurance policy. Your financial interests are then looked after by an insurance company.

You are choosing someone to look after your financial interests rather than doing so yourself. Retail financial "products" are really contracts for looking after your savings. Making everyone responsible for their own financial interests, implies a massive increase in private share ownership.

The Treasury's CAT standards only cover the first two layers of charges 1)-2) in Section 2 above. How can consumers be expected to cope with a hierarchy of charges of increasing stealth? The February 2000 occasional paper of the FSA states: "Investors appear to give explicit charges (let alone implicit costs) little thought." Its October 1999 consultation paper states: "So the reality is that while people might agree it is wise to shop around, relatively few people do." Charges should be the responsibility of the government, regulators and trustees, rather than individual investors.

The membership of occupational pension schemes has decreased in recent years. We need to introduce a new kind of mutuality, one which is similar to the Danish ATP system discussed below. The recent problems of Equitable Life result from a different kind of mutuality. It has to promote itself in the market by selling its policies, which does not apply to the ATP organisation because most membership is compulsory. Stakeholder pensions are compulsory in the sense that there are compulsory contributions, which pay either for national insurance, or for the stakeholder pension. They are expected to be mostly rebate-only. People will be told: "The government is giving money away. Sign here."

Part 3. Partnerships

9. "A partnership with financial service companies"

Consultation Brief 1 starts: "The Government is committed to encouraging more people to save for their retirement." This produces an incentive not to disclose the existence of hidden charges. It is not fair to encourage people to save, whilst at the same time allowing savings to be eaten-up by hidden charges.

The financial services industry also wants more people to save. In the green paper Partnership in Pensions, it states: "Stakeholder pension schemes will develop in a number of ways. All are likely to involve a partnership with financial service companies."

The DWP (DSS) refers to "the commercial financial services companies who provide stakeholder pension schemes". This implies that providers are commercial financial services companies. But stakeholder schemes could in theory be self-administered.

Stakeholder schemes can be set up by a) employers, b) membership organisations, c) financial service companies. The kind of employer willing to set up a self-administered stakeholder scheme, would probably already have an occupational scheme. Stakeholder pensions with providers are like extending group personal pensions to all employers with some modifications - especially the cap on charges. A membership organisation or employer could set up a self-administered scheme without providers, but has not so far done so.

Only trust-based stakeholder schemes can be self-administered. Stakeholder pension schemes may be established as either a trust or by deed poll. The former schemes have trustees and the latter "managers". OPRA states: "to look after the interests of their members, schemes must have either trustees or stakeholder managers". The managers are being referred to as "providers". If there are no trustees then your interests are looked after by providers, that is commercial financial services companies.

OPRA does not approve of this arrangement. In its response to the green paper it states: "For stakeholder pension schemes to be - and feel - secure, they need to be significantly different from existing personal pensions.":

"In Opra's view the only safe alternative to trusteeship for adequate stewardship of a stakeholder pension scheme would be to have an independent board of management and prescribed rules. This would be similar to the pension schemes in European countries (eg the Netherlands) whose legal systems are not underpinned by trust law. Some UK public sector occupational schemes have this type of structure."

A stakeholder scheme with managers is defined as contracts between the manager and individuals in regulation 2(1): "one or more contracts to be entered into between the manager of the scheme and each member of the scheme". The Post Office stakeholder scheme is "managed by" Standard Life. Some people will think they are depositing their savings with the Post Office rather than Standard Life! If the manager is an insurance company then the stakeholder pension is legally an insurance policy. The company owns the underlying assets and has its name on share certificates. Strictly speaking, the provider is not "managing your fund" because the funds do not belong to you.

The TUC has no doubt good intentions in setting up its stakeholder scheme with the Prudential. But since employers do not have to make contributions. It seems to some extent just like helping the Prudential sell insurance policies. There seems to be a partnership between financial service companies and not only the government, but also trade unions. The printers' PIPS stakeholder scheme and the AEEU stakeholder scheme are not registered with OPRA. MSF is promoting Axa insurance policies.

A stakeholder pension which is an insurance policy invests in insurance funds which are part of the life fund of insurance companies. They do not provide accounts to the public like unit trusts. Thus the funds of most stakeholder schemes will not have proper accounts. A stakeholder scheme with trustees has to have audited accounts, but not a scheme with managers.

10. The Danish ATP scheme

Stakeholder pensions are based on a partnership between the government and the financial services industry. They are therefore organised for the benefit of the government and the industry rather than for the benefit of members. The Danish Labour Market Supplementary Pension (Arbejdsmarkedets Tillaegspension ATP) is a fully funded pension scheme, which is compulsory for employees. It is based on a partnership between employers and employees. The members of the scheme are part of the partnership - unlike most stakeholder schemes. B&CE Insurance (Section 4.3 above) is a mutual company apparently run by employer and employee representatives, like the ATP organisation.

"ATP is an independent, self-supporting institution founded in 1964 and operated jointly by the parties to the labour market."

From the point of view of limiting costs, stakeholder pensions are based on the wrong partnership. The Director's Report for the year 2000 states that administrative expenses were DKr 111m (£ 1 = DKr 11.86 on 13th August 2001):

"Pension activity costs amounted to DKK 85m of this amount (equivalent to DKK 20 for each member), whereas investment activity costs accounted for DKK 26m (equivalent to 0.011 per cent of assets or just DKK 6 for each member)."

The previous Directors Report states that it is strenthening its asset management with the intention that "in 3-5 years' time it will be at the level of the best internationally".

There is an expense ratio of 1.4 per cent in the accounts, which is expenses as a percentage of contributions rather than capital per annum. The providers of stakeholder pensions have been arguing that even 1 per cent per annum of capital is inadequate! In the United States the funds of the Teachers Insurance and Annuity Association all have TERs less than 0.5 per cent.

The share portfolio turnover has been increasing in recent years:

Table 3

Share portfolio turnover of ATP scheme

year	1998	1999	2000
a) total end year DKr billion	74.1	99.6	109.9*
b) purchases DKr million	6,215	15,822	29,117
c) sales DKr million	5,171	10,633	25,557
turnover 100 x (b + c)/(2 x a)%	7.7	14.3	24.9

*52,260m domestic, 57,610m foreign

The ATP accounts seem to make no mention of the late payment of contributions which is the main problem reported by OPRA. From the point of view of OPRA the main duty of the trustees and managers of stakeholder schemes seems to be to ensure contributions are paid on time. OPRA has a considerable list of problems with occupational pension schemes, such as late audited accounts and failure to appoint advisers, none of which seem to occur with the ATP scheme.

In addition to impressive Annnual Report and Accounts available (in English) on the internet, the ATP scheme is superior to stakeholder schemes in other respects. It has a properly appointed board, regional meetings and a magazine for members. It has established a internet service: "Pensionsinfo makes it easier for individuals to obtain a complete overview of his/her own pension rights." We need a similar organisation in this country. Stakeholder schemes when launched in April, will not even have accounts.

ATP describes itself as a co-operative owned by the members. Since membership is compulsory for employees, it does not have to promote itself in the market to survive. Stakeholder pensions are sometimes seen as a rival to occupational pensions. The PensionService of the ATP organisation on the contrary, helps to administer several occupational schemes.

In the ATP 1998 accounts the Special Pension Saving Scheme (SP) administered by ATP, is stated to be run by only three staff. Contributions are "1% of incomes of people aged 16 to 66 - wage earners, self-employed . . . ". On retirement:

"Balances are paid out by monthly instalments over 10 years from a person's 65th birthday. If a member of SP dies before all the instalments have been paid, the balance of his account will be paid to his estate."

This is clearly building on the ATP scheme, which is run by 136 staff. In this country the government comes up with a saving scheme, like stakeholder pensions, which is introduced after extensive consultation with the industry, to be sold by an army of financial advisers and salesmen. The Danish system is highly superior.

The concern about small percentages of the explicit charge of stakeholder pensions, whilst ignoring hidden charges is a sad contrast with ATP. For example Toby Walne writes, in the last of his recent series of articles on stakeholder pensions in The Financial Mail on Sunday, concluding on 1st April:

"Some providers . . . plan to shave a bit off this one per cent maximum . . . This difference might not seem like much to begin with, but over many years, even a 0.1 per cent reduction in charges can save customers thousands of pounds."

Part 4. Size and stability

11. Economies of scale

The Danish ATP organisation has been growing in importance, ever since it was founded in 1964. With 4.2 million members, a reason for its efficiency is economies of scale. Charges which are a percentage of capital, like the maximum 1 per cent charge of stakeholder pensions, do not allow for economies of scale. A comparison of expense charges for funds of different size is shown below, from a survey of international funds by Timberlake & Company. It can be seen that the TER is smaller for the larger investment trusts, but not for the larger unit trusts.

Table 4

The cost of investment and unit trusts by size of fund

TER = total expense ratio, M = management charge, No = number of funds in survey

fund size £ million	investment trusts			unit trusts
	TER	M	Number	TER	M	Number
0-30	1.37	0.73	5	1.62	1.29	25
30-1000	0.99	0.73	7	1.64	1.39	21
100-300	0.62	0.45	9	1.28	1.10	12
300-1bn	0.48	0.45	11	1.68	1.56	4
over £1 bn	0.32	0.31	5

The managers of these investment trusts tend to charge whatever it costs to run the trust, so they pass on economies of scale, which does not happen for the unit trusts which charge a percentage of capital. The one per cent cap on charges does not encourage economies of scale.

A recent OECD report Private Pension Systems: Administrative Costs and Reforms discusses the administrative cost of funded pension schemes in OECD countries and Latin America, and in particular the Premium Reserve system in Sweden, for which contributions are a mandatory 2.5 per cent of earnings. This has restrictions on fund charges which depend on the size of fund (page 108).

Stakeholder pensions are intended for employers who do not have pension schemes, which tend to be smaller employers. Large pension schemes with many members are more stable than small pension schemes. Occupational pension schemes are wound up when an employer goes out of business, which is of course more likely with small employers. The 1997-98 Annual Report of OPRA, states that almost 15 % of all occupational pension schemes are currently being wound-up, namely 17,000-23,000 schemes.

There are many small schemes as can be seen from the 1999-2000 Annual Report.

Table 5

The size of pension schemes

Size of Scheme (number of members)	number of live schemes	as % of total live schemes	number of members	number of members as % of total members
2 to 11	94,800*	80%	293,000	1%
12 to 99	16,400	14%	592,000	3%
100 to 999	6,200	5%	1,859,000	8%
1,000 to 4,999	1,100	1%	2,399,000	10%
5,000 to 9,999	200	-	1,457,000	6%
More than 10,000	300	-	16,770,000	72%
Total	119,000	100%	23,370,000	100%
(this figure includes live public sector schemes)

*This figure includes 3,000 schemes where the Pension Schemes Registry knows only the number of active members

The Myners Review has a section on the training of the trustees of occupational pension schemes.

"The review recommends that funds and their sponsors should increase their investment in training for trustees."

This is a problem because trustees are so numerous. It is rather similar to the program of public education in financial services of the FSA, to help people make informed investment decisions. Most people are too busy.

OPRA reports that the number of small occupational pension schemes is declining and number of large schemes increasing.

"The number of occupational pension schemes (excluding public service schemes) dropped by over 11,000 in 1999/2000. The following table shows the change in numbers by size of scheme. It was interesting to note that the number of large schemes – those with over 1000 actually increased by 5%."

There is a similar trend in the United States, as reported by the PBGC.

" Today there are nearly 40,000 private-sector defined benefit plans insured by PBGC. This compares with a high of about 114,000 in 1985 The decrease has been primarily among plans with 100 or fewer participants, while the number of large plans insured by PBGC - those with 1,000 participants or more - has shown modest growth. These large plans are responsible for the steady rise in the number of workers and retirees covered by PBGC pension insurance, going from about 38 million in 1985 to 43 million today."

In France employees have to be members of the AGIRC and/or ARRCO pension schemes for white and blue collar workers. The latter is a recent amalgamation:

"ARRCO brings together 93 supplementary pension organisations, which in turn are combined into 45 schemes covering the entire private sector. On 1 January 1999, this organisational structure will be streamlined to create a single scheme."

In short there seems to be a world trend towards larger pension schemes.

12. The need for stability

The development of stakeholder pensions has been influenced by 401(k) schemes in the United States. A complaint about these schemes has been the frequent changes in the regulations. The present website started with a change in the DWP (DSS) Guidelines, and there have already been several such changes. A booklet Partnership in Pensions A summary, received from the DSS last year states, for example, that stakeholder pensions are "secure - independent trustees will ensure that the rights and interests of scheme members are put first;", "flexible - if a member does not contribute for a period, no administrative costs will be levied to avoid existing savings being eaten up."

None of the changes in the Guidelines have apparently been to the advantage of future members. Most stakeholder schemes will not now have trustees. Savings will be eaten up by administrative costs, irrespective of whether or not there are contributions. There has been a consulation process which could be described as paved with good intentions.

Stakeholder pensions form a market. But markets are inherently unstable. A provider may be taken over by another provider. A company may demutualise. Money Management reports that "of the 238 (life) companies operating at 31 December 1970, only 59 were still operating in the life market under the same name at 31 December 2000" (March 2001, page 111).

The regulatory system is also unstable. AFBD, FIMBRA, LAUTRO, TSA and the SIB have disappeared. IMRO, PIA, SFA, have changed from being autonomous "self-regulating" organisations. (The PIA is not yet on the internet.) They are now part of the FSA, or at least have "service agreements". They will probably become self-regulating autonomous departments within the FSA, and firms will still be "regulated by IMRO" etc. These autonomous departments may in due course be located away from Canary Wharf. In short we could be back where we started.

13. Conclusion

This website started with details of new DWP (DSS) Guidelines, including information about further charges, not included in the one per cent cap. Almost everyone has been under the misapprehension that "stakeholder charges are capped at one per cent". Anyone who has already bought a stakeholder pension, thinking that charges are capped at one per cent, has been the victim of mis-selling and may like to have their money back.

Charges are coming to light which investors had not previously realised existed. Poor investment performance is often incorrectly attributed to wrong choice of investments, when the real cause is hidden charges. What are the hidden charges 4) in Section 2 above? How large are they relative to dealing costs? These topics need to be investigated, perhaps as part of the review proposed in the Myners Review:

"The Review recommends that the Government should initiate a separate independent review of capital and information flows around personal investment products."

Such a further Review is welcome, and has recently been set up under Ron Sandler. He has issued a consultation document to which the present author is writing a response

14. Recommendations

14.1 Stakeholder schemes

1. It needs to be made clear in all publicity for stakeholder pensions, that in addition to the maximum 1 per cent charge there are other charges, especially for dealing.

2. There should be a limit on the proportion of the share portfolio of a stakeholder fund, which can be sold/bought according to whether the fund is growing/declining.

3. There needs to be an investigation to find out what are the other hidden costs 4) (Section 2), beyond dealing costs.

4. All stakeholder schemes and the funds in which they are invested need to have audited accounts showing the details in Section 6.3.

14.2 The long term

Introduce a national scheme organised on a regional basis which: provides a pension after retirement, has properly appointed trustees, meetings for members, central offices to represent members on a national level probably in London, Edinburgh and Cardiff, proper accounts and so on.

This national scheme must be run by a new organisation rather than existing financial institutions. Membership will need to be compulsory to some extent, such as for employees who are not members of occupational pension schemes. Contributions should be quite small at least initially to get the scheme started.

Any comments on this website will be gratefully received.