2.2 Stockbrokers' commissions

The FSA states: "As with all charges, commission costs affect investment returns." (1.2), but as the Fitzrovia Report states:

"Transaction costs in the UK are currently 'hidden' and the issue needs to be raised so that it can be addressed and, potentially, so costs can come under greater control." (page 232)

"Even for a company that specialises in investment research, it is currently difficult to analyse the trading costs without having disclosure guidelines in place." (page 228)

Trading costs are hidden because, as described in the Sandler Review, they are "netted off investment return" (10.137). The FSA states that: "Fund managers already keep records of commission and taxes." (97) They could therefore be disclosed. When a fund manager buys and sells shares through a stockbroker, the prices are generally net of commission, that is they are taken out of the capital of savers.

The management charge and expenses of unit trusts and OEICS can now be paid out of capital rather than income, especially following "Part C CHARGING EXPENSES TO CAPITAL" of FSA Consultation Paper 32. ^x

The FSA holds consultations and then follows the wishes of the majority. This sound very democratic, but it results in it doing the bidding of the industry:

"The eleven responses to CP 14 were overwhelmingly in favour of the proposal to allow fund managers to charge expenses to capital in whatever proportion they regard as appropriate ... The main area of disagreement was the need for a meeting of holders to give permission to the manager to change the way expenses are charged. " (3.3)

Taking charges out of capital increases the yield. The FSA states:

"Reaching for high yield products can be a dangerous strategy, which could put consumers’ capital at risk."

It also conceals charges since they are no longer reflected in the size of the dividends paid by funds.

The first of the conclusions of the Treasury Select Committee in its report on split capital investment trusts is:

"We believe that transparency in all aspects of the charges borne by investors should be paramount." (page 33)

In the Consultation Paper of the FSA Bundled Brokerage and Soft Commission Arrangements (April 2003) it states:

"Increasing the transparency of commission costs will improve the information available to consumers about the costs of investing. This will enable them to assess more accurately the benefits and risks of investing in managed funds and thus to make better-informed and more suitable choices." (A 2.13)

But this report does not recommend that commission costs should be disclosed by being included in the accounts of unit trusts and OEICS. But it says:

"Our proposals may also lead to changes in disclosure requirements...This could be part of normal fund accounting arrangements." (4.40)

This seems vague. Which "disclosure requirements"? Which "fund accounting arrangements?

"Our proposals will benefit consumers by ensuring that fund managers acting on their behalf have stronger incentives to obtain value for money," (1.10)

The FSA proposes that soft commissions should be abolished but not for investment research. Soft commissions will be limited to investment reseach so that commissions will be "limited to the purchase of trade execution and of investment research". ^x The amount spent on investment research might expand so that total amount of soft commission is the same!

The FSA says that "we intend to review the governance of retail funds authorised unit trusts and OEICS..." so that they can benefit from enhanced disclosure of soft commissions.

"We recently said in our Policy Statement to our CP185 consultation that we intend to review the governance of retail funds authorised unit trusts and OEICS, investment trusts marketed through savings schemes and managed funds of life assurers. We believe this could help to deliver the benefits of enhanced disclosure to private investors. They are unlikely to have the knowledge or market power to engage directly with fund managers on these issues." ^x

The Policy Statement CP185 does indeed refer to "the further work we intend to undertake on the governance of retail funds" (2.37) What work? Paragraph 1.8 says "retain the current governance structure" for retail funds. ^x

In the US and Canada ^x information about stockbrokers' commissions is certainly not as readily available as turnover ratios. Nevertheless savers are advised to consider commissions:

"One ongoing expense that is not included in the expense ratio is the brokerage costs incurred by a fund as it buys and sells securities. These costs are listed separately in a fund's annual report, sometimes as a percentage and sometimes as a dollar amount ... For example, Dreyfus Fund and USAA Growth & Income are both have reasonable expense ratios, 0.71 % and 0.89 % respectively. However, if you consider the brokerage costs of the funds, it turns out that Dreyfus Fund, with annual brokerage costs of 0.50 % is a more expensive option than USAA Growth & Income, which incurs only 0.09 % in brokerage costs." ^x

The portfolio transaction costs, that is dealing costs, of mutual funds in the US have to be disclosed under the Mutual Fund Integrity and Fee Transparency Act (2003), which requires the development of rules by the SEC and more recently the Mutual Fund Reform Act (2004). ^{x x} At the time of writing (February 2004) this is a Bill rather than an Act since it has apparently not yet been enacted.

New mutual funds regulations are resulting in the US from the late trading scandal. For example an article in Business Week reports that:

"The SEC has made clear it wants mutual funds to give investors more information about fees, portfolio turnover rates, revenue-sharing payments to brokerage firms, and the compensation of portfolio managers."

This is all a contrast to the UK where the government and FSA go out of their way to help the industry hide dealing costs.

The Financial Supervision Authority in Finland reports that in Europe:

"The costs that commissions cause the fund and its unit holders are due to the frequency of fund asset transactions, also known as the portfolio turnover rate.

In recent years there has been animated discussion in many European countries about commission-related fund costs. Although mutual funds, as referred to in the directive on collective investment undertakings (the UCITS directive), can be marketed freely between member states in the European Economic Area, the member states do not have a harmonised practice for reporting to investors on the impact of the commissions."

On the Plexus website dealing charges are represented as an iceberg. There is at least a tip showing above the surface. In the case of retail savers in the UK, this iceberg is entirely submerged. Dealing costs are outside the 1 % cap on charges for CAT standards, stakeholder pensions, the proposed "stakeholder products". and are not included in the FSA's Comparative Tables. They are mentioned in the Sandler Review (7.27). But it does not seem to use the term "portfolio turnover" at all. It has an emphasis on "products" as does the green paper on pensions Simplicity, security and choice: Working and saving for retirement (December 2002):

"Stakeholder pensions developed the idea of simple, easy-to-understand products," (page 82, paragraph 24) "The Sandler review's proposals builds on the ideas behind the stakeholder pension." (page 85, paragraph 39)

Dealing (or transaction) costs are concealed from purchasers of these "products". However the government does think dealing costs are important for occupational pension schemes and should be understood by trustees. It says on the Treasury website:

"Transaction costs are an important cost to pension funds. For trustees to fulfil their duty to act in the best interests of their beneficiaries, trustees must ensure that these costs are properly managed."

"Despite a great deal of controversy over Mr Myners’ proposals on commissions, few responses to the Government’s recent consultation disagreed with his central proposition, that these costs 'which are substantial' are 'subject to insufficient scrutiny' and that 'clearer and more rigorous disciplines could be applied'. On the contrary, the clear message from the consultation responses is that the problem is, if anything, greater than Mr Myners originally suggested ... Trustees, or those to whom they have delegated the task, should have a full understanding of the transaction-related costs they incur, including commissions."

Dealing costs are also important for savers but they do not have trustees to look after these costs which are mainly hidden. Savers are not even being informed that they are important. They are being treated as second class citizens, in comparison to the trustees of occupational pension schemes. The government is moreover promoting stakeholder schemes which generally do not have trustees.

There are over 100 thousand occupational pension schemes, which all have trustees. Having "a full understanding of transaction-related costs" may be rather too much to ask of so many busy people working on a voluntary basis, who are required to cope with the complexity of pension regulations, FRS 17, etc. Judging from Tables 7 and 8 below, trustees have allowed the portfolio turnover of pension funds to become excessively high.

2.3 Stamp duty

In the case of unit trusts, there are two kinds of stamp duty, referred to in the CAT standards for ISAs as a) "stamp duty on transactions in fund assets" (paragraph 37) and b) "stamp duty in dealings in units" (paragraph 38).

The amount of a) never seems to be given in the accounts of unit trusts. It is like a stealth tax. The amount of b) is sometimes given. It is known as "stamp duty reserve tax (SDRT)".

The CAT standards for ISAs say that a) is outside the 1% cap on charges and b) is inside. This seems to be different from the cap on charges for stakeholder pensions, judging from the OPRA website which refers to stamp duty reserve tax being outside the cap.

2.4 Purchases and sales

I like the presentation of the accounts of J O Hambro Capital Management, giving numbers of shares bought and sold, the price and date. For example 160,000 KVAERNER INDUSTRIER A shares were bought for the European Fund on 19th December 2001 at a cost of £ 99,200. Of these shares 60,000 were sold the next day on 20th December for £ 37,600. 60, 000 x 99,200 / 160,000 = 37,200. There was a £ 400 loss.

In the case of Hambro's UK Growth Fund, I do not understand why there are so many shares starting with FIRSTGROUP PLC sold on 9th November 2001, which it bought on 6th November 2001 for the same price.

These accounts only give the main purchases and sales. There seems no reason why all purchases and sales with numbers of shares, prices and dates, could not be made available. This would help students of fund management. Such publication seems likely to discourage high turnover.

It can be seen from the accounts of unit trusts and OEICS that asset managers are quite often buying and selling shares in the same company. This is another kind of portfolio turnover, which can be exactly defined as a percentage. If the numbers of shares bought and sold, with the price and dates were disclosed, then unit holders would be able to study whether they benefit from this practice. As it says in the Fitzrovia Report:

"Transparency at all levels generates the oxygen of trust." (page 7)

3. Disclosure in comparative tables

The turnover ratios in the Fitzrovia Report should be publish on a website so that savers can compare the turnover of different funds.

As discussed on my website Comparative Tables: The Hidden Charges the FSA is unwilling to included dealing costs in its Comparative Tables.

According to an article in the Financial Times (7th April 2003) FSA U-turn on fund performance data the FSA is doing a U-turn about the inclusion of performance data in its Comparative Tables. It will hopefully do another U-turn about the inclusion of portfolio turnover and dealing charges.

A reason given by the FSA for not including a measure of dealing costs in the Comparative Tables, is that they are hard to project forward in time. An analogy is fund managers not informing the trustees of occupational pension schemes about dealing costs because they are hard to project forward.

4. What are the dealing costs resulting from portfolio turnover?

The Fitzrovia Report states:

"It is not the intention of this report to analyse the cost of portfolio trading at fund level. Information on dealing costs is not in the public domain and some companies do not have the systems in place to monitor such costs. We are however taking a first step to exploring the impact of dealing costs within a collective fund by looking at the trading levels." (page 1)

Dealing costs need to be added to the other charges to obtain the total cost of saving. They can be estimated by taking a multiple of the turnover ratio. For UK equity funds the Fitzrovia Report quotes the 1.8 % of Kevin James, mentioned on my website Comparative Tables: The Hidden Charges. The average portfolio turnover for equity funds in the Fitzrovia Report (page 13) is 67 %, giving a dealing cost of 0.67 x 1.8 % = 1.2 % per annum of capital. The FSA gave a recent estimate of dealing costs of 1.5 - 1.8 % (Investors Chronicle 14 - 20 March 2003 page 88) for 100 % turnover.

The Sandler Review states that "the average unit trust underperforms the market by 2.5 per cent a year" (Foreword and Summary. No reference is given. In the home page, the figure of 3.7 per cent for the underperformance of unit trusts was quoted.). About half the underperformance is caused by dealing costs, and the rest by explicit charges.

Some accounts or prospectuses specify that the managers have soft commission agreements, such as Dresdner RCM mentioned above, and Bedlam Asset Management mentioned below. Soft commissions are payments for services performed by the broker. This seems to be mainly research, that is recommendations on the choice of investments. These services are only performed if there is a specified amount of trading. The Fitzrovia Report does not specify which funds use brokers which have soft commission arrangements. This could be in the next edition. Soft commissions are one possible reason for high turnover. Fitzrovia states:

"There can be an issue if the fund manager is comfortable with the existing portfolio, but has to trade as a result of the 'soft commission' arrangement." (page 228)

The July 2002 report Institutional Investment in the UK by Paul Myners recommended that in the case of occupational pension schemes, soft commissions should no longer be permitted, because they are not a transparent method of payment. They are charged directly to the capital of the fund. What do management fees pay for if it is not research? Which shares to trade could be decided entirely by the stockbroker.

Philip Augur, author of the well known book The Death of Gentlemanly Capitalism (2000), in an article on soft commissions Time to abolish tabloid equity research in the Financial Times (19th February 2003, page 17) states:

"Much brokers' research is closer to advertising than independent analysis. ... Unbundling research and execution would be an important step towards clearing the decks for an assault on the next great issue: proprietary trading."

"Proprietary trading" refers to trading by in-house stockbrokers.

The Sandler Review complains about the "bias in favour of active fund management" (page 135). There would be less of a bias if dealing costs and portfolio turnover were disclosed. Dealing costs are determined by portfolio turnover in the first instance, rather than whether or not a fund is actively managed. There are also explicit charges. The Review states:

"The charges of UK unit trust tracker funds, which have only minimal differences between them, vary from as low as 0.3 per cent per year to as high as 2 per cent per year." (page 1, paragraph 5)

5. Portfolio turnover of unit trusts, investment trusts, pension funds and insurance companies

The government publication Financial Statistics gives total assets, sales and purchases for unit trusts/OEICS, investment trusts, pension funds and insurance companies. It defines "turnover" (Table 5.3A) to be total purchases plus total sales. Dividing by total assets, we define:

turnover for the year % = 100 x (total purchases + total sales for the year) / ( total at end of previous year + total at end of year).

This turnover is calculated in Tables 7 and 8.

The turnover levels for overseas shares shown in Table 8 are substantially higher than for UK shares shown in Table 7. It would be interesting to extend these tables backwards in time from 1993. The turnover levels for insurance companies and pension funds seems to be increasing. The ratios of 52.9 % and 93.3 % for the UK and overseas shares of pension funds in 2000 seem remarkably high:

"Back in the 1980s it was accepted wisdom that pension funds had long-term liabilities and so had a matching policy of holding shares for the long term. Then the average was five years. Now it is two."

The turnover of occupational pension schemes seems remarkably high in the UK - comparable with unit trusts. In Canada:

"Portfolio turnover ratios are lower in pension funds than among other institutional investors and the size of individual holdings can be very large. These factors lead pension fund managers to encourage investee companies to improve shareholder value rather than simply to dispose of their shares.

Unlike the mutual fund industry, pension funds are not competitive. Their focus on performance is one of ensuring returns adequate to meet anticipated liabilities, not to attract investors."

Investment "to attract investors" is discussed by Kevin James in his CSFI report Waiting for Ariadne (2002):

"Funds act to maximize profits. A fund's profit increases as its revenues increase; a fund's revenue is the product of its funds under management and its (total) expense ratio. Hence, profit maximization implies that funds will act to maximize expected total revenue. Given that investors behave as described above, how would a fund go about doing so?

Consider a young fund competing for the business of the typical retail investor. Since new investors pick funds that beat the market by substantial amounts and existing investors display large amounts of inertia, the fund has a strong incentive try and beat the market through active management - even if such management on average detracts from performance. If it gets lucky, new investment pours in; if it gets unlucky, it's not much worse off." (page 14)

Turnover on the London Stock Exchange increased from £ 610 billion in 1990 to £ 5,581 billion in 2001 (Oxford Research Associates (OXERA) report An Assessment of Soft Commission Arrangements and Bundled Brokerage Services in the UK (April 2003), paragraph 102). This cannot have been caused by an increase in investment opportunities. It must have resulted in an increase in dealing costs for many pension funds. This has been causing concern such as to the trades union Unison reported in an article in the Guardian (2 March 2002):

"Fees charged by City investment banks to pension funds for buying and selling shares are costing the schemes an estimated £ 8 bn a year."

The London Stock Exchange states that UK equity turnover in 2002 was £ 1,815,034.2 million. Multiply by 1.8 % the cost of a trade in Table 11 below. Divide by 58.8 million people in the UK. In 2002 the cost per person in the UK of trading on the LSE in UK company shares is therefore:

£ 1,815 x 0.018 / 58.8 = £ 556

Table 2.1 of the OXERA report states that the total funds under management in the UK at end 1999 was £ 2,857 bn. If average costs and charges are 2.06 % of capital per annum, this is equivalent to charges of £ 1000 per person per annum for all the 59 million population of the UK.

Ned Cazalet estimates the expenditure of life assurers on administration and sales to be £ 570 per family per annum. Adding in dealing costs, unit trusts, oeics, and pension fund management, then this £ 1000 per person per annum seems a reasonable estimate.

6. Total dealing costs in the UK

6.1 Calculated from stockbrokers' commission

In their response to the Fitzrovia Report, Barclays Global Investors states:

"The combination of crossing and best execution makes our trading costs very low, resulting in far less drag on performance."

"Crossing" is trading directly between principals without going through markets, so there is no buy/sell spread. This should clearly be encouraged. Perhaps in the next edition of the Report, it could be stated whether or not funds use crossing networks.

The February 2000 Occasional Paper of the FSA The Price of Retail Investing in the UK by Kevin R. James, quoted in the Fitzrovia Report, estimates "the cost of a round trip trade" to be 1.8 % of the capital traded (Table 2, page 23):

Table 11

The Cost of a Round-Trip Trade ( 100 % Turnover )

Commission	0.30 %
Buy/Sell Spread	0.75 %
Price Impact	0.25 %
Stamp Duty	0.50 %
Total	1.80 %

These costs are borne by the saver. Who benefits? In the case of stamp duty it is the government. At least it spends the revenue on hopefully worthy causes. In the case of the buy/sell spread the beneficiaries are market makers in the City. This is a method individual day traders use to make money. It is possible to set up a more generous buy/sell spread. Suppose the existing buy/sell spread is x/y. A day trader can offer a buy-sell spread z/w where x < z < w < y.

It can be seen from Table 9 that the cost to savers of the buy-sell spread is substantial. Funds should be required to say whether they use crossing networks. If funds are buying through the market without using crossing networks the buy/sell spread should ideally be recorded and the total given in accounts.

According to the Inland Revenue stamp duty was £ 3,711.3 bn in 1999-2000, £ 4,476.9 bn in 2000-2001. If stamp duty is 0.5 % and commission 0.3 % as in Table 9, then total commission for 1999-2000 is £ 3.7 x 0.3/0.5 bn = £ 2.2 bn.

As quoted above the London Stock Exchange states that UK equity turnover in 2002 was £ 1,815,034.2 million. Multiplying by 0.5 % we obtain stamp duty £ 1,815 x 0.005 billion = £ 9.1 billion. The City must be under-paying stamp duty!

The cost of fund management has been discussed recently in two articles in the Times (14 February 2003) and one in the Financial Times (15/16 February 2003) .

"Mr Myners ... urged investment managers to justify the £ 2.5 billion cost of actively trading UK equity portfolios each year.... He said nearly £ 1 billion of the trading costs was paid as commission to investment brokers. A further £ 700 million is thought to be paid in stamp duty to the Treasury, with the rest spent on custody fees and other expenses. ."

Which "other expenses" in addition to stockbrokers' commission? Custody fees are not included in Table 9. From the Inland Revenue statistics for stamp duty £ 700 million seems much too small. In the Consultation Paper of the FSA Bundled Brokerage and Soft Commission Arrangements (April 2003) it states:

"In 2000, for example, UK fund managers paid about £ 2.3 bn from their clients' funds to UK brokers." (1.1)

The report from OXERA An Assessment of Soft Commission Arrangements and Bundled Brokerage Services in the UK (April 2003) says that total brokerage commission increased "from £ 1.5 bn in 1992 to £ 5.7 bn in 2000" (paragraph 103) Of the £ 5.7 bn, £ 4.5 bn was paid by institutional clients as opposed to private clients (paragraph 218). Payments (£ bn) to UK brokers for the year 2000 are therefore:

It would be interesting to know how much is being paid by UK fund managers to brokers abroad. The total amount paid by fund managers "as commission to investment brokers" could be £ 4 bn rather than the £ 1 in the article.

The commission calculated in the OXERA report agrees with the 0.3 % in Table 9 (slightly less). If commission is £ 4 bn we then obtain total dealing costs for UK institutions in 2000 = £ 4 x 1.8 / 0.3 bn = £ 24 bn.

The OXERA report says that the total institutional funds under management in the UK, at end 1999 is £ 2,477 bn (Table 2.1). Therefore 1 % of these funds are being spent each year on dealing costs.

6.2 Calculated from portfolio turnover

From Tables 5, 6 and 9, total shares traded in 2000 (£ bn):

	UK	(135 + 27 + 255 + 347) =	764
	overseas	(163 + 35 + 168 + 264) =	630
	total		1394

Multiply by 1.8 % and divide by 2 to get a round-trip trade.
Total cost of dealing by UK institutions in 2000 = £ 1394 x 1.8 / 2 bn = £ 12.5 bn.

This is substantially less than the £ 24 billion calculated above from the commission.

6.3 Calculated from total assets

The same £ 2.5 bn per annum estimate for dealing costs is contained in the Financial Times article of 15/16 February. But then in the same article it says:

"Elkins-McShery, a consultancy, estimates pension funds spend an average 0.27 per cent of assets under management on transaction costs a year and they absorb a further 0.2 per cent on market impact costs."

Total assets are given in Table 12.

Table 12

Assets of institutions (£ billion)

year (end)	unit trusts	investment trusts	insurance companies	pension funds
1993	88	29	434	480
1998	163	47	776	699
2000	223	60	933	765

Source: Financial Statistics , HMSO, April 1966, April 2000, September 2002.

Multiply transaction costs 0.27 % by year 2000 total assets of pension funds, we obtain £ 765 x 0.27 % bn = £ 2.1 bn.

The total assets at end 2000 were £ (223 + 60 + 933 + 765) bn = £ 1981 bn. Therefore "transaction costs" are £ 1981 x 0.27 % bn = £ 5.3 bn. This is approximate since the turnover varies between sectors as can be seen from Table 7. In the article the 0.27 % only applied to pension funds. There is a further £ 1981 x 0.2 % bn = £ 4.0 bn for market impact costs, and a further amount for the buy/sell spread. We again find the cost to savers of dealing by fund managers is substantially higher than the £ 2.5 billion of Paul Myners.

The FSA estimates that the dealing costs of stakeholder pensions could be 1.3 % of capital per annum. This is about the same as the 1.2 % for the dealing costs of equity funds mentioned in Section 4 above. It is about half the 2.5 % underperformance of unit trusts in comparison to the market mentioned in the Sandler Review (page iii).

If the 1.3 % applies to the £ 1981 total assets, then dealing costs are £ 1981 x 1.3 % bn = £ 25.8 bn per annum. This is about the same as the £ 24 bn calculated above from the total commission.

In Excessive Portfolio Turnover there is an estimate of £ 20 bn per annum for the cost of active fund management. The £ 20 bn is an estimate of Kevin James. It is slightly out-of-date and does not include pension funds. £ 25 bn is a better estimate. Someone reading the articles in the Times might think that the total cost to UK investors of trading by fund managers is £ 2.5 bn per annum, excluding explicit charges. But this is wrong by a factor of ten!

6.4 Justification of dealing costs

The Times reported that Dick Saunders, chief executive of the Investment Management Association, said:

"A fund manager should be ready and able to justify his turnover of the portfolio."

How can he do this? By giving all purchases and all sales of shares, with dates and quantities of shares and prices, as discussed in Section 2.3 above.

The article in the Financial Times mentions that:

"Under a new code, fund managers will have to detail trading commissions, costs and volumes to pension funds."

They should also have to detail commissions to unit holders of unit trusts and OEICs and insurance funds, personal and stakeholder pensions.

7. Sandler stakeholder products - a charter for churners

The watchword of the green paper on pensions is "simplicity". For example:

"Chapter 5 describes how the Government is working with the financial services industry to empower consumers to make the right savings choices by providing simpler products." (page 45, paragraph 50)

"The Sandler Review concluded, based on a rigorous analysis of the way competition works in the markets, that a price cap would be an essential component of the design of stakeholder products. . . There are also strong inherent attractions in a simple 1 per cent annual management charge." (page 87, paragraph 50)

"Attractions" to whom? Before 1979 there was a 0.5 per cent cap on the annual management charge of unit trusts - which is half. To be exact there was a maximum 13.5 per cent over 20 years. So there could be an initial 3.5 per cent and then 0.5 per cent per annum. The stakeholder 1 per cent takes no account of economies of scale.

The government states:

"The Government is committed to encouraging more people to save for their retirement." (paragraph 10)

There is a slippery slope from encouraging saving, through the hard sell to mis-selling. The government has already slipped some way down this slope, because the stakeholder "cap on charges" does not include all charges as discussed on my website Stakeholder Pensions: The Hidden Charges. People are buying stakeholder pensions without being made aware of the dealing costs.

The Sander Review complains about "the proliferation of products". Its proposed "stakeholder products" seem like a further proliferation. Surely in the first instance we should find products which are no longer needed, such as arguably endowment policies for mortgage protection. On average they do not seem to have benefited policyholders and therefore should not have existed. ^{x x} A survey of closed with profits funds by John Chapman in Money Management (September 2003, page 31) discussing endowment shortfalls says:

"Even by year 10 some companies were projecting surrender values well below the £6,000 premiums paid on £50 pm policies - Britannic £3,400, R&SA £3,350, Sun Life of Canada £3,540 etc. Such institutionalised exploitation has been removed with pension plans, but it has remained with endowments."

The Review says that the 5 per cent withdrawal rule is unsatisfactory (paragraph 8.51). I agree. Abolishing this rule would be a step in the direction of reducing the number of "products".

The Treasury's Consultation Document Proposed product specifications for Sandler "stakeholder" products (February 2003) also stresses simplicity, for example:

"Simplicity. Few in the target market will have significant understanding of financial products. ... Simple "stakeholder" products will therefore appeal to this target market." (47)

The "target market" is people on "low to medium incomes".

Does the "charge cap" for stakeholder products include the "additional expenses" referred to in the Sandler Review : "which include trustees' fees, registrars' costs, custody expenses, audit fees and regulators' fees" (7.27)? According to the stakeholder regulations dealing charges "incurred directly in the sales or purchase of securities" are outside the 1 % cap. I have misgivings about how strictly this is enforced. There are grey areas. For example custody fees might count as a dealing charge, since they tend to increase with more dealing. They are included in the "cost of actively trading UK equity portfolios" estimate of Paul Myners reported in the Times (above Section 6.1). They might therefore be outside the cap.

There is the same problem of grey areas and enforcement with the reduction in yield (RIY) indicator. It does not include dealing costs, and these are not the only costs which it does not include. The rules for the calculation of the RIY are tightened up from time to time. This implies that savers have been misled by RIY values smaller than they should have been. For example in the Consultation Paper of the FSA Informing consumers: product disclosure at the point of sale (February 2003), it says:

"We intend to tighten up the existing COB rules to prevent mutual insurance life companies from offsetting surplus from non-profit business against charges and expenses in order to disclose a smaller RIY." (5.73)

According to the Sandler Review dealing charges will be outside the cap for the new stakeholder products. This implies that simplicity is an illusion. It does not seem clear from the Review which other charges will be outside the cap. There are likely to be grey areas.

The Sandler Review also proposes a "stakeholder with-profits" fund:

"The provider would also produce an annual report setting out the financial condition of the fund, asset allocation, investment performance, and costs charged to the fund." (10.136)

But it is once again proposed that dealing costs should be concealed:

"There would be clear regulatory prescription on the treatment of costs, to ensure consistency across providers and products. In principle, this should say that all costs should be shown, rather than being netted off investment return. However it will be important to ensure consistency with unit trusts/oeics, which have certain costs which can be treated in this way." (10.137)

This is referring to section 8.5.4 of the regulations for Collective Investment Schemes.

The Occasional Paper of the FSA To switch or not to switch that's the question An analysis of the potential gains of switching pension provider (2002) by Isaac Alfon, states that personal pensions "tend to have high portfolio turnover and high undisclosed charges":

"Switching between personal pensions is between broadly similar managed (or balanced) funds, ... These tend to have high portfolio turnover and high undisclosed charges. ... Switching between a personal pension and stakeholder pension may well be, in effect, a switch between a managed fund and an index-tracker." (page 16, footnote 20)

How many purchasers of personal pensions realise that there are high undisclosed charges? They could be even higher in the case of Sandler stakeholder products. But hidden charges are not mentioned in the list of warnings for purchasers of these "products" in the Sandler Review (paragraph 10.24).

The undisclosed charges are not included in the stakeholder cap on charges. The need for this cap implies that the industry cannot be trusted not to make excessive explicit charges. It can arguably be even less trusted not to make hidden charges. The FSA says that portfolio turnover should remain undisclosed (Section 2.1 above). We have to remain content with statements such as: "tend to have high portfolio turnover and high undisclosed charges".

What proportion of stakeholder pensions are index-trackers? Perhaps there is a "bias in favour of active management" which is a heading in the Sandler Review (page 135), which reports that in 2001, 93 % of the new units of unit trusts were for actively managed funds.

Without knowing dealing charges it is difficult to determine whether to switch or not. The FSA said that the cost of dealing, additional to the stakeholder 1 % cap on charges, "could add up to" 1.3 % as quoted in the Sunday Times. This suggests that the average turnover ratio for stakeholder pensions is about the same as the average 67 % for equity funds in the Fitzrovia Report.

The data in the Fitzrovia Report is compiled from published accounts. The proposed new "stakeholder products" will not necessarily have accounts. The Treasury's Consultation Document Proposed product specifications for Sandler "stakeholder" products states:

"From a consumer's perspective, a mutual fund and a unitised life product are essentially the same product" (6)

Insurance funds do not have separate accounts like unit trusts. All unitised funds with units bought by the public, should have proper accounts. Most purchasers of stakeholder products may not have time to read them, but this is no reason why they should not be available. There may be the illusion of simplicity, but without proper accounts stakeholder products cannot be trusted.

8. Looking after other people's money

The industry is defined by government legislation, regulations and regulation. It can be thought of as a manifestation of this legislation, regulations and regulation. Any problems in the industry implies that there is something wrong with these. Various problems are discussed in the Sandler Review. This is largely concerned with "products". It complains about the "proliferation of products" such as the bullet points in paragraph 3.13:

opaque and inconsistent terminology;

lack of clarity and consistency in report of product charges;

proliferation of products; and

product differentiation that does not reflect true differences in what is being offered.

On the other hand several publications of the FSA stress the importance of not restricting "product innovation".

Some of the terminology in the Sandler Review seems itself to be rather opaque. For example it uses the term "mis-selling" without defining what it means, and then says "The Review believes it would be helpful if there were greater clarity about what constitutes mis-selling." (147) If the Review is unclear about the meaning of mis-selling what hope have the rest of us, at least those of us who are not lawyers?

I have been sent the following definition of mis-selling:

"Try Section 47 of the old Financial Services Act (1986). Now Section 397 of the Financial Services and Markets Act (2000).

The heading in Halsbury's Statutes , admittedly, possibly not part of the Statute itself, is entitled:-

'Misleading information as to investments'

And Wp insurance policies are legally included under investments.

What more is required?"

Mis-selling is not defined in the FSA's glossary of terms, and the FSA says that "Arguably, mis-selling is not a regulatory concept at all.". Other terms used frequently in the Sandler Review, such as "product regulation" and "regulated product" are also not defined in the glossary of terms or in the Review. As in "substituting product regulation for the current regulation of advice" (10.13). This substitution might not be very effective. But then the Review says: "Assessing the effectiveness of the regulatory regime is not within the remit of this Review." (5.1) The Review does not define what it means by "effectiveness" or even "product". (Is for example an investment trust a "product"? It is referred to as "a type of product" in paragraph 97.) Nor does it define what is "the industry", and so on.

Some of the most important problems affecting savers in the UK at present are:

1. Increase in the number of complaints

"Last year, the Financial Ombudsman Service handled a total of 462,340 initial enquiries from consumers and received a total of 62,170 new complaints. This year, the Financial Ombudsman Service expects to receive 98,000 complaints, an increase of 58% on the previous year." ^{x x}

"This year" is 2003-4. Half of these complaints are for mortgage endowments.

2. Savings become trapped

Savings become trapped in closed funds ^{x x} and in "products" such as: with-profits annuities, ^x with-profits bonds, ^x personal pensions. ^x

3. Funds are passed between companies

Such as the sale of Royal & SunAlliance’s closed life business to Resolution Life. ^x

4. Continuing mis-selling

Example:

"One of the most shameful things I was told to do as a financial adviser was to obtain a signature of a 75-year-old woman on a form where it was clearly stated that she felt sound of mind and body and had no need to have a friend or relative present to help her make her decision to invest £34,000 to provide for her disabled son when she died. Actually, I had heard her quite clearly state that she wished her son-in-law were present."

5. Poor investment performance

For example: "Not a single unit-linked endowment would have made you money over the past ten years and very few pensions break even."

Investment performance has been particularly poor for funds which have become closed (see e.g.: 1. 2. 3. 4. )

6. Excessive charges

"It is impossible to exaggerate how deeply flawed were the operations of the life and pensions industry in the 1990s. ... The industry used an exploitative model with high up-front charges. Allied Dunbar was the exemplar, with a 5 per cent initial charge, a 1 per cent annual charge, a monthly policy fee and initial allocation of premiums as low as 35 per cent for 2 years. Even Standard Life had a 5 per cent initial charge and a 95 per cent allocation rate, so that 10 per cent of your premiums disappeared. With 40 per cent of pension plans lapsing in the first 4 years, one in three pension savers lost money." ^x

"My company put in 5% of salary, I put in 5% of salary and the pension company took out 90% in charges." ^x

7. Excessive complexity

For example look at the 209 (at the time of writing) Consultation Papers of the FSA. Some of the reports on the industry are of extreme complexity, such as the Baird report on Equitable Life.

8. Instability

Companies are constantly being taken over or amalgamating with other companies. For example Money Management reported 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).

In the US John Bogle complains about:

"Soaring Turnover Among Mutual Funds. Fifty years ago, there were 126 equity mutual funds, most with prudent long-term investment objectives. Today there are 4,800 equity funds... During the 1990's, 55% of funds - more than one half! - failed to survive."

9. Inadequate disclosure of information

On this web page we are discussing disclosure of portfolio turnover and dealing charges.

10. Deteriorating standards of service

For example: "People approaching retirement can no longer assume insurers will pay out benefits on time."

11. Large orphan funds and unclaimed assets

Unclaimed assets could arguably be reduced if national insurance numbers were specified on all life insurance policy documents, and if they were associated with accounts such as bank accounts. People (or beneficiaries if deceased) could then be traced through the National Insurance Contributions Office.

Foreign companies help themselves to orphan funds which should belong to UK policyholders - such as AMP in the case of Pearl ^{x x x} NPI and London Life. ^x

12. Regulatory failure

This is discussed at enormous length on the Motley Fool discussion board for Equitable Life (example 1, example 2), where other problems are also discussed (example), in a general atmosphere of betrayed trust.

The government wants people to "engage with financial institutions". This is a reason for the introduction of Child Trust Funds. ^x In view of the above problems it can be argued that the less people engage with financial institutions the better. There is a difference between short term saving for a rainy day and long term saving such as towards a pension. Everyone should have capital for an emergency but they should not be burdened with responsibility for long-term financial investments. There are too many problems.

These problems result especially from long-term saving being looked after by companies as a business. The financial industry is constantly saying that restrictive regulation is harmful to the industry and London as a financial centre. There is a conflict in the government objectives of wanting to have a successful industry and wanting people to benefit from saving. The government has to resort to various "incentives", to encourage the financial industry to provide better value for money for savers, such as the stakeholder cap on charges and the proposal to abolish soft commissions.

Businesses are formed to look after people's savings, and the expenses of the business are charged to savers' capital (example below).

In the pensions green paper there is a contrast between the "1 % annual management charge", and the average cost of occupational pension schemes in the UK:

"for the largest schemes (those with over 40,000 members), the average administration cost is around £ 28 per member a year. For smaller schemes (those with fewer than 500 members), the cost is around £ 80 per member a year." (page 129, paragraph 10)

The switch away from occupational pension schemes to group personal and stakeholder schemes (discussed in the home page), results in a substantial increase in costs.

The administrative cost is less for very large schemes. For example the Danish ATP scheme has 4.3 million members. In the year 2002:

"Pension activity expenses amounted to DKK 27 for each member, while investment activity expenses were DKK 16 per member."

A charge for administration of (DKK 27) about £ 2 per annum is quite a contrast to the stakeholder 1 % of capital per annum! Total costs and charges in the UK of the part of the industry concerned with fund management, that is life assurance companies, fund management companies and stockbrokers, is about £ 1000 per person - as estimated in Section 5 above.

The ATP scheme is compulsory for employees. Contributions are a fixed percentage of salary. Employee contributions to the Thrift Savings Plan for Federal employees in the US are optional.

"Your agency will start to contribute Agency Automatic (1%) Contributions to your account. If you are contributing your own money, you will also receive Agency Matching Contributions."

Charges are not so low as the ATP scheme, but still far lower than the stakeholder 1%. TERs are below 0.1 %:

"For example, the G Fund expense ratio was .06 percent in 2002. This means that in 2002, earnings in the G Fund were reduced approximately $0.60 for every $1,000 of G Fund account balance."

The worry for savers is that the task of looking after their savings is an excessively profitable business, or alternatively that promises will not be kept, as has happened in the case of the Equitable guaranteed annuity rate (GAR) policies. A Central Discontinuance Fund (CDF) is suggested in the pensions green paper (page 65). It could become an organisation in which pensioners could have confidence. It is advocated by the TUC. It would provide pensions for members of occupational pension schemes which are wound-up. It is funded by such schemes.

In its paper Action on occupational pensions (June 2003) the government proposes to set up a "Pension Protection Fund":

"We will establish a compensation scheme known as a Pension Protection Fund, run by a statutory body, to protect private sector defined benefit scheme members whose firms become insolvent with unfunded liabilities in their pension scheme." (7)

This is not a CDF fund as discussed for example in a paper of the Institute of Actuaries. This would provide pensions for members of defined benefit schemes which are discontinued, that is wound-up.

"Would take over the assets and liabilities of defined benefit schemes winding up where the sponsoring employer has gone out of business and could therefore provide no further support." (2.1)

A firm need not be "insolvent with unfunded liabilities". Solvency insurance is discussed as an option in the paper (9.14).

A CDF might be expanded in due course to look after personal savings, and perhaps provide pensions for the employees of companies which do not have an occupational pension scheme.

Various people advocate index-tracking as opposed to active fund management. The more important question is whether or not savings are being looked after as a business, or by trustees. For example occupational pension schemes are provided by employers. But they are not in the business of providing occupational pension schemes. They have some other kind of business. Group personal pensions are provided by insurance companies which do so as part of their business.

Trustees may employ fund management companies. But occupational pension schemes are institutional investors, as opposed to individuals who are retail savers. I prefer the term "saver" in a unit trust or OEIC, rather than "investor". I do not like frequent use of the term "consumer" as in Treasury's Consultation Document, which does not use the term "saver" at all. "Investor" and "consumer" seem to have the same meaning in this publication. "Consumers for these products" (paragraph 44), is using "consumers" in the same sense as consumers of hamburgers! An investor is someone who invests in a specific business, rather than just putting money aside for a rainy day.

Large savers in unit trusts and OEICs are able to negotiate reduced charges in various kinds of "share classes", as mentioned on the Fitzrovia website:

"There is a rapid movement by many fund management companies towards the creation of new share classes. This allows the segmentation of retail and institutional investors. Institutional classes have lower fees and a better track record of performance." (click on "Latest News" and "Trends and Developments in Annual Fund Charges")

This shows the advantage of saving on a group basis rather than on an individual basis. A second example is the reduced market value adjustment (MVA) negotiated with Equitable Life by group schemes in comparison to that applied to individual policyholders. A third example is group AVCs in comparison to free-standing AVCs:

"The average return on an in-house AVC is 10 to 15 per cent. higher than for a free-standing AVC." ^x

In conclusion saving should not be looked after as a business, and it should be on a group basis rather than on an individual basis.

9. Suggestions for future editions of the Fitzrovia Report

The Fitzrovia Report is a first edition. There are various ways it could be expanded.

I would like to see it contain data on all UK-based unit trusts and OEICs, which is about 2000 rather than the 1,341 in the Report. It should also be extended for example, to personal and stakeholder pensions.

It could be interesting to know whether the fund is a unit trust or an OEIC.

The start and end year assets would be a helpful information.

There could be information about dealing, such as whether or not there are soft commission agreements, and whether crossing networks are used.

10. Recommendations

This website has two general themes, corresponding to specific recomendations about the disclosure of portfolio turnover and expense charges, and a general recommendation to establish a new institution to look after savings on a trust basis. Recommendations are:

Specific:

1. The following should be put in the annual accounts of unit trusts and OEICS:

- total stockbrokers' commission for the year,

- total stamp duty,

- portfolio turnover ratio,

- all purchases and sales of shares, with price and numbers of shares and dates. It would be sufficient to put this information on the internet.

2. Turnover ratios should be available in comparative tables on the internet.

General:

3. There should be a move to a new financial system in which savings are looked after on a trust basis, rather than as a business.

11. Conclusion

The Fitzrovia Report states:

"High or low portfolio turnover is not necessarily good or bad." (page 10)

There needs to be a study relating the turnover statistics in the Report to investment performance, which would probably determine whether high turnover is good or bad for savers, one way or the other. Fitzrovia have not expressed the intention to do this study, although the Report says:

"The effect that portfolio turnover, and hence transaction costs, have on fund performance should be a consideration for investors and asset managers alike." (page 7)

Such studies have been done in the US. For example:

"Trading costs, like expense ratios, are negatively related to fund returns and we find no evidence that on average trading costs are recovered in higher gross fund returns ... the striking result is the remarkably close association between total fund costs and fund returns."

The Fitzrovia Report seems to be in favour of low turnover. There is a table on page 227 showing "Performance Drag" for different levels of turnover. E.g. 100% turnover results in a 1.7% "Performance Drag" - including stamp duty. The Report is very diplomatic. It says for example:

"Low portfolio turnover levels can benefit both the investor and the asset management company." (page 7)

"Collective funds is an area where you do not necessarily receive better service, more comfort or a superior experience by paying more for it." (page 226)

Portfolio turnover as discussed in the Fitzrovia Report, is a substantial extra complication for savers. People who like such considerations should be investing directly in shares rather than in unit trusts and OEICS. Savers should not need to understand and negotiate expense charges at all, or as little as possible. This should be the responsibility of trustees. There is a need for a new financial system with new institutions.

It seems that our savings are being drained away by entirely unnecessary and excessive dealing. Fund managers are selling to other fund managers in a merry-go-round which does not benefit savers.

Appendix

An example of OEIC charges

1. "No gain - no fee"

Charges can be illustrated with reference to the prospectus of a new OEIC Bedlam Funds plc. It has a "no gain - no fee" method of charging. Its charges in other respects are typical.

An article in the Evening Standard reported that Bedlam Asset Management plc claims that savers are being "fleeced":

"SANDWICHBOARD men proclaiming 'They're bleeding you dry' and 'You're being fleeced' will patrol outside Merrill Lynch, Schroders and other top City investment houses next week as part of a knocking campaign orchestrated by a rival fund manager"

On its website it states:

"The fee charged will be 1.4 % per quarter on the fund's value, subject to the fee not reducing the gain in any quarter to below 1.25 %."

In a rising market 1.4 % becomes 1.4 x 4 = 5.6 % a year. It was explained in an e-mail:

"A gain of 5 % per quarter will accumulate to about 20 % per annum of which we are entitled to the management fee (5.6 %). And that is our fee structure - No upfront Charges, No Bid-Offer Spread, we don't Run a Box and We have no Exit Charges,"

The "management fee" is described as "total charges" in its marketing literature. Values are calculated over the last few years, based on the performance of the stock market. They are low because of a falling market. In a reasonably stable market they will average 1.4 x 2 = 2.8 %, due to fluctuations. The 1.25 % quoted above, will make little difference.

2. "Fees and expenses"

"Fees and expenses" will be paid by the three sub-funds (at the time of writing they contain only £ 100 each). The Explanatory Prospectus of Bedlam Funds states:

"All fees and expenses of the Company attributable specifically to .... any Sub-fund shall be borne out of that Sub-fund. Fees and expenses which are not attributable to any specific Sub-fund will be borne by each Sub-fund in proportion to their respective Net Asset Values." (page 18)

An addition to "management fees" there are:

"Administration Fees

... The Administrator's fee shall accrue at a rate of 0.2 per cent of the net asset value of the company as a whole and is subject to minimum of £ 82.5 thousand per annum. The Administor shall be reimbursed reasonable out of pocket expenses.

Custody Fees

The Custodian's fee shall accrue at the rate of 0.1 % of the Net Asset Value per annum .... The custodian shall be reimbursed reasonable out of pocket expanses including the fees of any sub-custodian appointed by it which will be charged at normal commercial rates.

Other Fees and Expenses

... including, without limitation:

taxes,
expenses for legal, auditing and consulting services,
promotional expenses,
registration and other fees due to supervisory authorities,
insurance,
interest,
brockerage costs,
costs of obtaining and maintaining a listing of Shares in any Sub-fund, and all professional and other fees and expenses incurred in connection therewith,
the cost of the publication of the Net Asset Value,
Directors' out of pocket expenses,
Director's fees, which are not expected in exceed £ 50,000 in aggregate per annum.

Preliminary Expenses:

The formation and preliminary expenses (including printing costs and legal fees) relating to the Company, are not material, and are estimated to be approximately £ 125,000. This sum shall be amortised in the accounts of the Company over its first five years and will be allocated between each sub-fund... " (page 17)

The founder of Bedlam Asset Management thinks that "the typical fund manager" does not do enough research:

"His modus operandi is using a pin and the back pages of the Financial Times and following the latest investment fad."

3. OEIC insolvency

Many unit trusts have converted to sub-funds of OEICs. The shareholders of OEICs do not have legal ownership of the underlying assets, like the unitholders of unit trusts:

"The body corporate becomes the beneficial owner of that money ... " (2.4.4)
"The property of the collective investment scheme must belong beneficially
to BC, " (2.5.3)

The sub-funds of OEICs are liable for the debts of other sub-funds, and those of the entire OEIC company. The prospectus of Bedlam Funds states:

"Although each Sub-fund will be treated as bearing its own liabilities, the Company will remain liable as a whole to any liabilities to third parties for all liabilities of the Company. However in the event of any insolvency of one or more Sub-funds under this umbrella structure, any creditors in respect of such insolvent Sub-fund or Sub-funds would be creditors of the Company as a whole and accordingly could proceed against any assets of the Company, including any assets held in other Sub-funds of the Company." (page 8)

Started January 2003,
Stephen Wynn.

E-mail: centre@boltblue.com

Any comments would be gratefully received. I have received the following:

28th March 2003

"I am delighted to see your notes on the website, what I saw as a stockbroker for some 35 years made me ashamed of my own cloth. Churning was one of my greatest irritants, and why I might have been a poor partner to have. I have strong views about them and their price fixing charges, but also the Unit trust, pension fund and other so called financial advisors. I have also been involved with many big and small Charities as a Trustee, and outspoken at some of the abuses I have seen, quite often from the most elite brokers and bankers."

Counter: