1. Most, perhaps all, of the advantages claimed for Personal Accounts such as "low-cost", were also claimed for stakeholder pensions.

2. The board of the Personal Accounts Delivery Authority (Advisory and Executive (3.8)) and the Personal Accounts Board should to a large extent be appointed by organisations other than the Government.

3. Choice of fund implies a choice of the fund management company providing the fund. There is a loss of flexibility, because it is difficult or impossible to change this company.

4. Choice will result in competition between funds which increases costs.

5. NPSS ¹ should provide income in retirement rather than obliging its members to buy annuities when they retire.

The White Paper says that Personal Accounts have a number of advantages compared with personal pensions and stakeholder pensions. Most, perhaps all, of these advantages, were already claimed for stakeholder pensions. Personal accounts are being referred to as a new kind of personal pension, such as on the personalaccounts.info website:

"They will effectively be personal pensions, run on a occupational basis and administered by employers.” ^x

Chapter 7 of the Green Paper A new contract for welfare Partnership in pensions (2002) ^x refers to the "complexity of personal pensions .. difficulties individual buyers face in getting a good deal .. costs of distributing and marketing pensions". But these difficulties and costs are still present. The White Paper refers to: "the complexity of the current pension system ... the relatively high charges associated with pension products" (1.3). The Green Paper says:

"Individuals can find personal pensions difficult to understand and nearly impossible to compare." (7)

"We believe this can be addressed for stakeholder pension schemes by defining a number of minimum standards which schemes must meet in order to be stakeholder pension schemes. These will promote standardisation across schemes, allowing consumers to compare what is on offer more easily. They will give a minimum level of protection and reassurance to potential scheme members that a stakeholder pension scheme will provide them with a good basic deal." (18)

"The introduction of minimum standards for stakeholder pension schemes will ensure a level of standardisation across schemes which will allow individuals to compare what is on offer more easily. Minimum standards will mean that people can have confidence that their pension is flexible and that any stakeholder pension scheme will give them a good basic deal." ^x

The White Paper says:

"There are currently 26 stakeholder providers on the Financial Services Authority list and it is very difficult for individuals to choose the right one for them." (2.11)

Stakeholder pensions are described as "flexible" in the Green Paper and White Paper (7.29). Personal Accounts are not described as "flexible" in the White Paper. There is a prohibition on "transfers between personal accounts and other pension products and schemes" (7.15). I am concerned about this, especially transfer values from occupational pension schemes. I discussed the topic of contributions to NPSS in my response to the white paper Security in Retirement (2006) (www.comparativetables.com/wpaper.htm#cont). ^x

2. "Run for its members"

The White Paper says that NPSS will be "run for its members and delivered by the private sector":

"We want an organisation that is independent of government, run for its members and delivered by the private sector." (3.1)

"harnessing the skills and expertise of the private sector" (3.2)

"It will be led by a chairman and chief executive, recruited almost certainly from the private sector." (3.7)

This is similar to the Green Paper:

"Stakeholder pension schemes need to be run with much greater emphasis on the interests of their members." (20)

"The reassurance provided by the minimum standards and by ensuring schemes are run in their members' interests will reduce the need for detailed financial advice when people join schemes." (22)

"Stakeholder pension schemes will develop in a number of ways. All are likely to involve a partnership with financial service companies. (64)

With so much involvement of the private sector, and with funds handed over to "fund managers", it seems to me that NPSS will be run to a considerable extent for the benefit of the private sector, rather than for its members. The "skills and expertise of the private sector" did not prevent the Equitable Life fiasco.

There is an example of running a financial company for the benefit of the company rather than its customers, in the disciplinary hearings of Equitable Life actuaries at the Institute of Actuaries: ^x The more policyholders with a guaranteed annuity rate (GAR) option exercise this option, the more the insurance company has to pay out. So that the company benefits by not drawing attention to this option:

"Whatever we do, however,there is a fine balance to strike between being open and not drawing an excessive degree of attention to the existence of GARs." (memo from Paul Headdon to Roy Ranson, paragraph 76, page 35) ^{x x}

The appointment of the chairman and chief executive of the Personal Accounts Delivery Authority (3.7) is clearly important. Which? comments:

“However, it is critical at this stage that the delivery authority continues to reflect a consumer focussed approach to personal accounts. This is best guaranteed if we see a true consumer advocate appointed." ^x

"A true consumer advocate" means someone for example from Which?, Investors' Association, United Kingdom Shareholders' Association, Equitable Members Action Group. These organisations are financed by consumers and investors. But representatives from Which? which was formerly the Consumers' Association, seem to be appointed only rarely to the boards of different bodies. Whereas representatives from the National Consumer Council, which is financed by government, are appointed by the Government to be the chairman or on boards of various bodies, ^x such as the chairman of the Financial Ombudsman Service, the board of the Financial Services Consumer Panel, Commission on Unclaimed Assets ^x and formerly the FSA.

The White Paper asks: "How can members’ interests best be represented in the governance of personal accounts?" (page 150). Do not hand over all the funds to fund management companies. Considering splitting up the default fund into a separate fund for Scotland and Wales, and perhaps even the English Regions in due course.

The Delivery Authority is "time-limited". It is only concerned with setting up NPSS. This could cause an accountability problem. The Personal Accounts Board will be able to blame problems on the Delivery Authority which no longer exists.

The White Paper says that NPSS will be "independent of government". But the "key leadership positions" (3.10) will be appointed by the Government. The Personal Accounts Board would be more independent of government if it was appointed by other organisations - like the board of the Caisse Nationale d’Assurance Vieillesse (CNAV) in France. ^x The board of the Arbejdsmarkedets Tillægspension (ATP) scheme in Denmark is specified in an Act ^x starting: "21.- (1) The board of representatives shall be composed of 15 employer representatives and 15 wage-earner representatives and a chairperson."

A contributor to a discussion commented:

"The scheme will give the Government investment control over an enormous consolidated pension savings fund which it ill no doubt use to prop up lame duck industries, fight wars, and spendon wasteful Government projects." ^x

This is not the case if NPSS funds are invested by independent fund management companies, or NPSS is itself properly independent from government. I prefer the latter approach. The White Paper says:

"It is vital to have independence from both politicians and pressure groups. This will be members’ money and their pension in retirement and it is vital that their savings are protected from external pressures." (5.12)

The Swedish Government seems to have a modest influence on the investment of the PPM default fund:

"Because of social policy considerations, the government prevents the default fund from investing in some well-known companies, including Coca-Cola, Exxon Mobil Corp., Liz Claiborne, Inc., and Sears Roebuck & Co." ^x

NPSS will need to be a politically weighty organisation able to stand up to external pressures.

3. Who will invest the default fund?

The "strategy for the investment of the default fund" is part of the "investment strategy". This is initially the responsibility of the Delivery Authority:

"For instance, the delivery authority with executive powers, is likely to be responsible for designing the investment strategy which will be in place when personal accounts are launched, whereas the personal accounts board is likely to be responsible for regularly reviewing that investment strategy." (3.15)

"The board will be responsible for ..:

investment strategy – ensuring that contributions are invested in the best interests of members; deciding on the range of choice available to investors; the strategy for investment of the default fund; and appointment and management of fund managers. (3.19)"

The default fund should not be "delivered by the private sector", but invested in-house. NPSS will have an “investment committee”. But it seems this will be concerned with “investment strategy” rather than “fund management” which will be “outsourced " (2.1).

NPSS is an "occupational defined contribution pension scheme" (6.19). Occupational pension schemes sometimes manage funds in-house. For example the 2006 Report and Accounts of the Universities Superannuation Scheme ^x (page 23) has three categories of investment:

"Total investments of the fund

Investments under the direct control of USS Ltd	58.7 %
Investments managed internally on the basis of external advice	12.3 %
Investments managed externally (passive)	1.2 %
Investments managed externally (active)	27.9 %"

The investments managed externally are mandates rather than funds chosen by the members.

4. "Fund managers"

The expression "fund management company" or "fund management companies" is not used in the White Paper. "Fund managers" is used instead, as in:

"These providers are generally linked to fund managers who would invest individuals’ funds." (2.27)

What happens if a particular fund managed by a fund management company performs poorly? Will it be possible to change the fund management company? This seems to be a problem where funds are chosen by NPSS members.

"Choice of funds" in the White Paper implies a choice of fund management company. This is not just "harnessing the skills and expertise of the private sector" (3.2), funds are also handed over to fund management companies. A choice of funds could still be provided investing funds in-house perhaps "on the basis of external advice". An example of such "external advice" seems to be: "St James Place contracts in fund managers to manage policyholders / investors money." ^x

5. Too many funds

The "NPSS model" proposed by the White Paper for the new system of Personal Accounts, is similar to the Swedish Premium Pension System (PPM). There is a "state-managed" or "quasi-state" default fund - the Premium Savings Fund part of the Seventh National Pension Fund (AP7) - and a choice of private sector funds. In an article about the PPM scheme, The Brookings Institution in the US reports:

"The default fund has outperformed the weighted average of actively placed funds in recent years, which may further have increased the attraction of non-choice." ^x

So what is the purpose of having more than the one fund - the default fund? The paper of the Investment Management Association Individual Retirement Accounts: International Evidence (2006) contains a table (Table 6) showing the numbers of new PPM members opting for the default fund dropped from 67% in the year 2000 to 9.4% in 2004. ^x The Regulatory Impact Assessment ^x of the White Paper says:

"Evidence from the Swedish PPM pension scheme suggests that the apparent benefits of choice can be illusory. In 2003 over 90 per cent of new members joining the scheme entered the default fund and only a small percentage made any changes to their portfolios. (2.29)

The implications of the research are that a single default fund is not likely to be appropriate for all types of saver" (2.30)

The "implications of the research" about the PPM scheme seems to me to be the exact opposite. A single default fund, with no choice of other funds, is likely to be appropriate for all types of saver. PPM has been criticised for having too many funds.

"Sweden’s Premium Pension System (PPM) fund range should be cut from 708 funds to just three simple risk-adjusted choices, according to plans proposed by the president of the AP7 state pension fund." ^x

The White Paper says that NPSS will "offer a limited choice of funds" (2.15). There are three categories of fund: 1) "a default fund", 2) "a small number of bulk-bought funds", 3) "a wider range of funds, which we expect to include social, environmental and ethical investments, and branded funds" (5.7). This could add up to be a considerable number of funds. A choice of funds is likely to result in lower overall returns, as discussed for example in the Australian article Super duper: more choice, lower returns. ^x

The default fund seems likely to become large. It should perhaps be split up in due course, such as a separate fund for Scotland and Wales.

6. Choice of funds

6.1 Marketing costs

“We know that the provider choice model has additional costs which are not in the NPSS model. Because firms compete directly for consumers, they have marketing costs which are not borne in the NPSS model.” (2.38)

But since the NPSS has a choice of funds, fund management companies will be competing for the contributions of NPSS members. Marketing costs were incurred in the case of the Swedish PPM scheme:

“Choosers in the initial round of fund choice in 2000 were exposed to approximately $65 million in advertising by fund companies.” ^x

"Increased choice leads to increased competition." (2.20) It also leads to higher costs.

"The Sandler review showed that there was little evidence that choice and competition in the pensions market drove down cost." (2.37)

6.2 "A significant minority"?

The White Paper says that most people do not want a choice of funds "overall no choice was generally preferred" (page 70). A choice of funds can be "complicated and confusing" (pages 70, 107). It produces complications such as switching between funds. Referring to investments the White Paper says: "Having a choice allows customers to switch." (page 71) The DWP Research Report No 370 said:

"As with provider choice, having a choice of funds was seen as making the scheme more complicated and confusing than it needs to be. The majority of participants in all groups – particularly the non savers – stated that this feature would result in them having to do a great deal of research to find out which fund would be best for them; which did not appeal to the majority of participants. As a consequence, they thought it likely that opt out rates would increase." (page 45) ^x

Only one fund was considered fairest:

"This was also seen as a way of making the scheme equitable. There was concern by some of the non-savers – particularly those earning between £5,000-£15,000 per annum – that if there was a choice of funds available, those with a greater sense of financial awareness would be able to profit while those who were new to pensions may play safe and, as such, have less income in retirement. Therefore, having one fund was seen by the majority to be the fairest – and easiest – solution that takes into account everyone’s needs." (page 46)

I expect the default fund will perform best. But it is in principle unfair if some funds do better than other funds. The same applies to different classes of policy, which has resulted in the Equitable Life arguments between the GAR and non-GAR policyholders.

The White Paper says that "a significant minority" want a choice of funds "perhaps up to 2 million " (2.29) out of "6 - 10 million" (5.7). The DWP Research Report says: "However, there were some groups that would like to be able to have a choice here." (page 46). But the minority who want a choice this does not seem to be quantified in the Report. So how can this be described as "up to 2 million" and "significant"?

An article in the Financial Times (19th February 2007) Watchdog warns design of national pensions default fund will be critical reported results from a survey of pension schemes:

"Strong evidence that most people will opt for the default fund comes from the NAPF's survey of existing workplace schemes, more than 80 per cent of which currently offer a default fund. Where they do, a mere 6 per cent of employees choose to invest in any of the other funds on offer." ^x

Six per cent is not "a significant minority". It is a small minority. This seems to be a more reliable estimate for the number of people wanting a choice of funds than the focus group discussions in the DWP Research Report No 370. ^x

A choice of funds may result in delays. In the case of the Swedish PPM scheme:

"There is a long lag in crediting pension contributions to private accounts. To save on administrative costs, contributions are credited to accounts once a year after the government has received income tax information. It takes on average 18 months from the beginning of the tax year for workers’ contributions to be credited to their account." ^x

There is an analogy with hospitals. Some patients may like a choice of hospital. But is it worthwhile building an expensive computer system to provide them with this choice? The majority of patients do not have an opinion beyond preferring the nearest. The initial choice of funds is the responsibility of the Delivery Authority:

"The delivery authority will finalise ... the investment strategy" (3.11).

"The delivery authority is likely to be responsible for: ... development of the investment strategy – agreeing the statement of investment principles, determining the level of choice available to members, designing the default fund and contracting with fund managers." (3.12)

Figure 5.1 in the White Paper shows participation rates falling for a wider choice of funds for 401(k) plans in the US. It says there is a reported increase in contributions for these plans: "people with a choice of investments contributed 8.5 per cent more into the scheme than those with no choice at all." (5.5). But it does not necessarily follow that people are contributing more as the result of having a choice. The White Paper quotes the paper by Papke L (2004) Choice and other determinants of employee contributions to defined contribution plans. This says that schemes for wealthier people are more likely to provide a choice of funds:

"Workers in service, farming and construction, and machine operator occupations are significantly less likely to have a choice than workers in managerial positions (10 to 24 percentage points)." (page 52)

Rather than "managerial positions" the "target group" for the NPSS model "tends to be younger and on low to moderate incomes" (1.1), with incomes shown in Figures 1.1 - 1.3. The White Paper proposes the NPSS should provide "structured choice ... a shortlist of funds .. for those that want it." (page 70). It seems that new members will be provided with a shortlist of funds whether they want it or not. They may then choose a fund from the list or do nothing.

6.3 A list of problems caused by a choice of funds

1) Only a small minority of NPSS members will want a choice of funds.

2) It is expensive, requiring the creation of the proposed Pension Payment System.

3) It is difficult or impossible to change the fund manager.

4) It will result in competition between funds to attract money, resulting in extra costs.

5) It presents people with an unhelpful dilemma about which fund to choose perhaps requiring financial advice.

6) There will be inadequate criteria for making choices, especially when the funds are new, that is without a track record.

7) It is catering for personal likes and dislikes, such as for or against index-tracking funds, which may not be based on much knowledge or experience.

8) However many choices are provided, some people may not like any of them.

9) It is storing up complaints in years and decades to come from those who (with hindsight) make the wrong choice.

10) Switching between funds will increase costs.

11) The number of funds is very likely to increase over time.

7. Loss of means-tested benefits

There have been discussions, whether or to what extent benefits received from NPSS will result in a loss of means-tested benefits. I signed the e-petition of the Prime Minister submitted by Steve Bee. ^x

The Minister, James Purnell says "the vast majority of people will be rewarded for voluntary saving." ^x There could be a minority who are not rewarded. So that a minority could be disadvantaged for the sake of the majority. This topic occurs for example, in Lord Neill's inquiry "regarding how complainants felt they had been dealt with by the Financial Ombudsman’s Service and whether those perceptions were well founded or not". ^x In the Opinion ^x he quotes the Ombudsman:

“In considering a claim, we think that the FOS would aim to be fair to all policyholders.” (128)

Lord Neill asks:

“Why should the FOS 'Aim to be fair to all policyholders'? Does a judge have to be 'fair' to all the other shareholders when adjudicating on a claim by one shareholder against a company?” (page 45)

Another example is allowing Equitable Life to trade after the House of Lords verdict. Sir Callum McCarthy, Chairman of the FSA says in an open letter to the Minister Ruth Kelly MP:

"The prospective benefit to its 1 million existing policyholders of a sale of the Equitable was bound to outweigh the prospective detriment to the 6,000 new policyholders who joined after the House of Lords verdict." ^x

8. Index-tracking

The White Paper refers to the US Thrift Savings Plan (page 95) as an example of low charges.

"Charges in the Thrift Savings Plan started at around 0.35 per cent in 1988 and reduced to 0.07 per cent by 2003. Thereafter, the charges have reduced by 0.01 per cent every year."

It does not point out that all the equity funds of this scheme are index-tracking funds. ^x It does not suggest that any of the NPSS funds should be index-tracking funds, and only seems to mention this topic once. Comparing active and passive fund management the White Paper says:

"Evidence to date suggests that both types of fund manager achieve a similar rate of return." (page 92)

On the contrary, I agree with The Motley Fool website that index-trackers achieve a better return on average:

"Whether index trackers are better than managed funds is the cause of a fair amount of controversy in the world of investment. The evidence is fairly clearcut however, and it shows index trackers beat the vast majority of managed funds over the long term." ^x

The advantage of index-tracking seems to be that it reduces portfolio turnover. It follows that it would be possible to have the same advantage by specifying that portfolio turnover must not exceed a number such as 20 per cent. This is a buy-and-hold strategy, on which there is an extensive literature, such as:

"A buy-and-hold strategy over a 44-year investment period turns out to perform as well as a strategy of investing in a broad market index. At the end of 44 years, the buy-and-hold portfolio tracks the S&P 500 index much the same as portfolios of most active managers. Although it is concentrated, its fundamental characteristics are attractive. This strategy will be more tax-efficient than indexing." ^x

The White Paper does not discuss index-tracking. There seems to have a bias in favour of active management (Sandler Review ^x page 135) by fund management companies:

"Large numbers of retail investors who are not in a position to identify superior managers are nonetheless investing in actively managed funds. This is of concern because the average actively managed fund underperforms the market index after charges. " (7.49)

Although not discussed in the White Paper it seems likely that some of the funds will be index-trackers. This will be determined by the NPSS Executive Delivery Authority and Personal Accounts Board.

9. Charges

9.1 Categories of fund

There is a heading Importance of low charges on page 91. "The Government is determined that charges will be fair, especially for low earners." (4.13). There are three categories of fund 1) "a default fund", 2) "a small number of bulk bought funds at low charges", 3) "a wider range of funds .. " (5.7). It seems that category 3) will not have such low charges.

9.2 Hidden charges

The White Paper does not mention hidden charges, which are described are "implicit costs" in the First Report of the Pensions Commission and "undisclosed costs" in the Sandler Review. They will need to be monitored by the NPSS Delivery Authority. The Treasury says:

"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." ^x

It could be a considerable task to monitor these costs for all the NPSS funds. The White Paper says the "cost in large-scale pension schemes can be driven down over time." (page 95). It can also be driven up - as happened with stakeholder pensions.

9.3 "Reduction in charge levels"

Stakeholder pensions are quite often described as "low cost" and "low charges". ^x In the White Paper NPSS is described as "low cost" and "low charges", but stakeholder pensions are not described as "low charges" - only "low cost". The White Paper says:

"The work the Government has done indicates that, in comparison with existing pension products, personal accounts would achieve a substantial reduction in charge levels for our target group." (4.7)

Stakeholder pensions achieved a reduction in charges in comparison to personal pensions. But the cap on charges for stakeholder pension has increased from 1 per cent to 1.5 per cent for the first ten years. The FSA has lifted caps on fund charges, as discussed in a article in the Financial Times Investors protest at FSA lifting caps on fund charges ^x (1st December 2006).

10. Annuities

Using mortality tables PMA92 (C=2010) ^x we can calculate two “yields”: 1) mortality yield, 2) investment yield. The mortality yield is the annuity rate which would be obtained if annuity funds are invested at zero rate of interest (with no charges). This is 100/(expectation of life) % per annum. The investment yield is the investment return allowing for mortality at different ages. It is the rate of interest used for discounting annuity payments which produces the purchase price of the annuity.

It has been claimed that PMA92 is out-of-date and the expectations of life too low. Therefore the above table is recalculated adding 5 years to the ages when people die, that is with the mortality the same as PMA92 when five years younger.

age	years	50	55	60	65	70	74
expectation of life:	years	37.98	33.06	28.22	23.49	18.98	14.87
monthly income:	£	415	448	506	584	678	790
a) annuity rate:	% pa	4.98	5.38	6.07	7.01	8.14	9.48
b) mortality yield:	% pa	2.63	3.02	3.54	4.26	5.27	6.72
a) - b)	% pa	2.35	2.36	2.53	2.75	2.87	2.76
investment yield:	% pa	3.55	3.55	3.76	3.98	3.97	4.04

There is no mention in the White Paper of "ASP" or "drawdown". I understand that it is no longer necessary to buy an annuity at age 75 ^x because of new rules for ASPs. I signed an e-petition on this topic. ^{x x} The White Paper seems to say that there is such an obligation, for example:

"For those who do not make an annuity decision by the age of 75, it will be necessary for annuities to be purchased on behalf of individuals. The decision on the default annuity at age 75 for personal accounts is an area the executive delivery authority is better placed to consider." (5.21)

A commentator to a discussion also remarked that there is no reference to "drawdown" in the White Paper and said: "This Government may be trying to edge them out of the picture for these new products." ^x

Why should people have to do anything with their investments when they retire, rather than keep the capital intact and receive the investment income? The NPSS organisation should have its own insurance company to provide an income in retirement. PPM in Sweden provides “fund insurance” and “traditional insurance”:

“Fund insurance means that you have holdings in funds. Your pension payments will vary from year to year, depending on how the value of your funds fluctuates up and down. Your pension is paid out every month.

If you choose fund insurance, your PPM account will remain intact and your money will continue to be invested in the funds of your choosing. You can still monitor your investments and switch funds whenever you like. You can opt for traditional insurance later if you like.” ^x

“The alternative to fund insurance is called traditional insurance. If you opt for traditional insurance, PPM will sell your fund holdings and manage your money. You are guaranteed a fixed minimum amount every month.

If you opt for traditional insurance, your PPM account will still exist but your money will be managed by PPM.

You will receive interest on your money every year, a so-called bonus interest. This can vary depending on how well the capital is managed. The sum is determined every month and published on PPM's website. ” ^x

The White Paper has a heading Accessing pensions savings (decumulation) (page 108). Suppose there is a pension scheme which consists of a fund which pays pensions out of investment income. Over time members of the scheme die and are replaced by new members. The capital remains intact, so that there is not any decumulation. It seems beneficial for pensions to be paid as far as in this way out of investment income. This requires the accumulation of a capital surplus. It seems to me that NPSS should be allowed to accumulate such a surplus for paying pensions after retirement.

Treasury Select Committee currently (February 2007) has an inquiry into Unclaimed assets within the financial system. ^x A surpluses may result from unclaimed assets. Where institutions are associated with affinity groups, surpluses can belong to the group and stay within the institution rather than being distributed to charity. NPSS is associated with an affinity group, namely employees "who do not have access to a good employer pension scheme" (page 47).

11. Too much outsourcing

There is indeed a need for "a new way to save", especially because of "the impact of charges" (4.4). But I have misgivings about so much of NPSS being outsourced.

The White Paper mentions "lessons learned from previous government IT systems failures" (page 163). One of the lessons seems to be too much outsourcing. The development of large computer systems is hard to outsource because of the difficulty of specifying requirements.

"The lack of senior IT management experience within the civil service has been cited by many a National Audit Office report as the reason for the series of high-profile government IT failures over the past two decades.

One of the key reasons for this is that the IT profession and career ladder within the civil service was virtually wiped out by the wholesale outsourcing of central government departmental IT to the big services and consulting companies in the late 1980s and 1990s." ^x

NPSS will have outsourced "administration and fund management services" (7.4):

"The new body would be responsible for oversight of the new systems – collection, account administration and fund management – but would outsource these functions to the private sector." (2.1)

"The board will be responsible for employing its own staff: it is expected that this will not be a large organisation as most of its functions will be contracted out." (3.20)

It seems that the outsourced fund management includes not only the choice of funds but also the default fund, which should be invested in-house in the first instance. NPSS could set up its own fund management company.

In my opinion there should not be a choice of funds, especially not actively managed funds. This is the "NPSS minus" model (page 66). A choice of funds will result in lower average returns. It introduces all sorts of complications. There is a need for extra record-keeping and to negotiate charges with fund management companies.

Requiring people to buy annuities when they retire is another kind of outsourcing. NPSS should provide income in retirement for its members, so that it will be necessary for it to have its own insurance company, perhaps like the Swedish PPM scheme which provides "fund insurance" and "traditional insurance".