Skip to main content

Pensions Bill

Volume 703: debated on Monday 30 June 2008

My Lords, I beg to move that the House do now again resolve itself into Committee on this Bill.

Moved accordingly, and, on Question, Motion agreed to.

House in Committee accordingly.

[The LORD SPEAKER in the Chair.]

Clause 28 [Power of trustees to modify by resolution]:

84: Clause 28, page 13, line 21, after first “scheme” insert “—

(a) ”

The noble Lord said: I shall speak also to government Amendment No. 85. Clause 28 enables trustees, with the employer’s consent, to make changes to an occupational scheme so that it can be used for automatic enrolment under the employer duty. We want to do as much as possible to encourage employers to retain their existing generous pension provision, so we have tabled Amendments Nos. 84 and 85 to extend the modification power. The amendments enable trustees, with the employer’s consent, to change the rules of a scheme to increase the contribution rate and the basis of its calculation in order to satisfy the quality requirements for both occupational money purchase and certain hybrid schemes. As we have already discussed, money purchase schemes must require contributions of at least 8 per cent of qualifying earnings, of which a minimum of 3 per cent must be paid by the employer. We recognise that many schemes provide for higher contributions than the minimum default required under the reforms and that such schemes will satisfy the quality requirement without needing to make adjustments.

However, other schemes will not, so the trustees and the employer may want to make changes to the rules to meet the required standard. These amendments to Clause 28 will facilitate such changes. This may be particularly helpful to employers offering schemes with restrictive rules that might otherwise prevent such changes being made. Where the power under Clause 28 is used, employers will not be required to consult scheme members on such modifications.

However, we have ensured that the necessary safeguards are in place to protect members’ interests. Changes made under the proposed amended Clause 28 must increase contributions. Any other changes can be made using provisions elsewhere in pensions legislation and will be subject to the normal rules on consultation with scheme members. This is to discourage levelling-down and retain the existing protection for scheme members through the consultation requirements. I hope noble Lords will support the amendments. I beg to move.

We have debated some of the issues that lie behind these amendments already in Committee. The amendments are part of the Government’s denial of the problems that the definitions in the Bill will involve. They are happy to grant the trustees a power to alter the scheme rules to allow an employer to pay more or more often into a scheme, but they have not recognised that the problems the Bill will cause will operate in a different way. Employers will have to pick up the costs of any rule changes, either directly or indirectly, and Clause 28 correctly requires employer consent. But they will not give that consent unless there is some advantage to them, and it is not obvious what the advantage will be of increasing contributions under the Bill.

We will not oppose the amendments, but they do not address the real issues that will arise under the Bill of why employers would want to incur costs to amend their own scheme—not only by paying more in but by changing the rules and their payroll systems to meet the definitions we discussed on our previous Committee day in connection with the group of amendments under Amendment No. 60A. The difficulties are so great that it would be easier for employers not to amend their existing schemes but simply to let personal accounts take over. Levelling-down is the elephant in the Committee. The Government cannot wish it away with amendments like this.

To some extent, the Government have got their timing wrong. If we were clearer on exactly how they see this working, we would be happier. Like the noble Baroness, we do not oppose the amendments in principle, but it looks as though the Government have not got their ducks in a row, if I may put it that way.

I am surprised at the reaction to these modest provisions. The noble Baroness asked what would encourage employers to make changes under their schemes to comply with the auto-enrolment and quality requirements of the Bill. It is fairly obvious: they would want to have, as they do now, good quality provision for their employees. They see the competitive advantage in many cases of good scheme provision and they do not want unwittingly to fall foul of the rules relating to qualifying earnings.

Qualifying earnings underlie the amendments and other matters that we will discuss. The Government’s aim is to maximise the number of new savers and savings and to support existing pensions arrangements. We do not want to disrupt arrangements that offer good pension outcomes for savers. As noble Lords know, a number of stakeholders raised concerns about the practical arrangements for dealing with scheme qualification in relation to qualifying earnings. We discussed some of these when we debated amendments to Clauses 12 and 19.

Since then, the Minister of State for Pensions Reform and I have been talking to stakeholders about how we can make the test as simple as possible while still preserving the new minimum level of pensions savings. We focused on a suggestion by a group of four stakeholders—the ABI, NAPF, ICAEW and SPC—the essence of which is annual reconciliation, which we have also been thinking about for a while. A key objective for the Government is to avoid disruption for existing schemes so that good arrangements may continue to function beyond 2012 without unnecessary changes or additional administrative processes. More detailed discussions with key stakeholders are already planned, and I fully expect to be able to be clear about our approach on Report. I hope that that has reassured the noble Baroness and indeed the noble Lord, Lord Oakeshott.

That was most encouraging. The issue will not go away, as we said on the previous Committee day, and I am glad that the Minister has taken the Government’s thinking a little further forward. There is clearly some way to go, but I am grateful for what he has put on record.

On Question, amendment agreed to.

85: Clause 28, page 13, line 22, at end insert “, or

(b) by increasing the amount required to be paid in contributions, in order for the scheme to satisfy the requirements contained in section 19(1) or those requirements as modified under section 23(1)(a).( ) An increase under subsection (1)(b) may be made only—

(a) by increasing the amount of any contribution, directly or by modifying the basis on which it is calculated, or(b) by increasing the frequency of any contributions.”

On Question, amendment agreed to.

Clause 28, as amended, agreed to.

Clause 29 [Deduction of contributions]:

86: Clause 29, page 13, line 33, leave out “an occupational pension” and insert “a”

87: Clause 29, page 13, line 34, leave out “6(3) or” and insert “or 6(3), or of an occupational pension scheme in accordance with section”

88: Clause 29, page 13, line 36, at end insert “(in the case of an occupational pension scheme) or the provider of the scheme (in the case of a personal pension scheme)”

On Question, amendments agreed to.

89: Clause 29, page 13, line 36, at end insert—

“( ) Regulations prescribing arrangements for the purposes of section 3(2), 5(2), 6(3) or 8(2) may require the employer to make such a deduction or payment at any time on or after the date with effect from which the jobholder is to become an active member of a scheme under the arrangements.

The noble Lord said: Clauses 3, 5 and 6 require employers to enrol or re-enrol their jobholders to a pension scheme. Clause 29 then permits employers to deduct pension contributions from the individual’s pay and pass that money to a scheme. Amendment No. 89 clarifies that regulations prescribing the automatic enrolment process may require the employer to start making pension deductions from a jobholder’s salary from the first pay day after automatic enrolment. In practice, that means that if a jobholder is paid between the time they are automatically enrolled and the time when all the practical arrangements to get them into pension saving have been completed, the employer may be required to deduct pension contributions from their wages. That ensures that pension saving can start from the day a jobholder is automatically enrolled. I beg to move.

This is the fifth or sixth of a large group of government amendments that we are due to look at today. Many of them are eminently sensible, but as my noble friend Lady Noakes has just said, we have queries on some of them. Incidentally, by my count, the number of amendments that the Minister’s advisers hope we will get through by the end of the day is 77. I suspect that they are a little erroneous in that view, but we shall see.

Amendment No. 89 is supposed, as the Minister said, to allow the deductions to be made from the first pay day after the auto-enrolment. However, the amendment does not actually say that; it says,

“at any time on or after the date with effect from which the jobholder is to become an active member of a scheme under the arrangements”.

If it said what the Minister thought it said, I would be perfectly happy, but

“at any time on or after”

does not give me that confidence.

This is part of a continuing debate we are going to have about the structure of the Bill. It is about taking a power so that we can ensure that employers have to deduct upfront under any arrangements; otherwise, some might be tempted—although I accept that this is overwhelmingly not the case—to take contributions later on, even though there is an automatic enrolment requirement from day one. That would disadvantage employees in respect of their pension arrangements. This provides for a power so that the regulations can make clear when those deductions can and should take place and require that they take place upfront.

The noble Lord will be aware that the precise mechanics of auto-enrolment and the opt-out period are still being worked through, which is why the provision is perhaps not as prescriptive as he would like, but the intent is very clear: to prevent arrangements being put in place whereby contributions were taken from employees down the track, after they had been auto-enrolled, so that they did not get the full benefit of that auto-enrolment. I accept that the provision is not as prescriptive as the noble Lord would wish, but these powers are part of the structure of the Bill and enable what we want to happen. I think we have agreement on what we want to happen.

I fully understand what the Minister wants to happen, but I do not think, with the best will in the world, that the amendment achieves it. The Minister quite rightly wants the money to be taken on the first pay day after auto-enrolment—I accept that; it is absolutely right—but I advise him to withdraw the amendment and think it out again.

The amendment seems to be pretty straightforward. It states:

“Regulations prescribing arrangements for the purposes of section 3(2), 5(2), 6(3) or 8(2) may require the employer to make such a deduction or payment at any time”,

et cetera. The noble Lord may balk at the term “may”, rather than “should” or “will”, being used, but the amendment provides a power so that regulations can require the employer to make those deductions from the start. I am happy to ask officials to look at it again, but I believe that it is clear and does not need it.

Clause 29, as amended, agreed to.

89A: After Clause 29, insert the following new Clause—

“Means-tested benefits

(1) Before the coming into force of the other sections of Part 1 of this Act, the Secretary of State shall publish his projections of the numbers of people likely to be subject to means-tested benefits (including housing benefit) and in at risk groups following the introduction of personal accounts.

(2) If the projected figure published in accordance with subsection (1) exceeds 10% of the total expected pensioner population, the implementation of the scheme under section 58 of this Act shall be postponed for at least 12 months.

(3) The Secretary of State shall publish any proposals for reform to address the findings in subsection (1) above concurrently with the publication of the projections thereunder.”

The noble Lord said: The amendment is reasonably self-explanatory. It is about having a report on means-tested benefits, and would introduce a requirement for the Secretary of State to report on the Government’s projections of how the introduction of auto-enrolment into personal accounts—indeed, any account—will impact on those likely to be subject to means-tested benefits. It would postpone the implementation of the scheme if that projection were too high.

After pressure in another place, the Government have done quite a lot of work on the impact of means testing, and have recently taken steps to slow the growth in the number of pensioners it will affect. I understand that that work is continuing. These reforms to the state pension and the second state pension have had a large impact. Under the Government’s original scheme, we would have had nearly 80 per cent of pensioners on means-tested benefits in 2050. Under the new scheme, we are looking at an estimated 30 per cent or so still being eligible. This is a huge improvement, but the number is still high. I am optimistic that there is more improvement to be made. This is not the place for a debate on how this could be achieved, but acknowledgement of the figures is crucial to the success of auto-enrolment, and the question of how to increase the amount of pension saving in the country.

The underlying assumption that enrolment into a pension scheme is a good thing underpins the whole Bill. From the central implementation of auto-enrolment to the provision of generic advice, the Government will be sending a message loud and clear to the target market that it is a good thing for them to put money into a pension scheme. Even before we get to that point, I believe that the Government need to run a lead-in campaign, probably along the lines of “Tell Sid”, which was so successful in persuading people to buy into the privatisation of British Gas in the 1980s. Both messages need to be unambiguous as they will be directed at those who do not have a deep understanding of the issues that surround pension provision and are not placed to assess any government advice against their own circumstances. It is therefore even more important to ensure that this message is not misplaced.

I know that the Minister will fully agree with me—this time, anyway—on the importance of ensuring that qualifying schemes are seen to be worth while for all those participating. The Government have produced reports and presentations seeking to reassure us that the vast majority of contributors will have a return on their savings and that those who will not could never have been predicted in any way. Some of these studies are still in the pipeline and will further inform the decisions on the final form that personal accounts and qualifying schemes will take.

If the Government are serious about targeting personal accounts effectively and are careful about the accuracy of their generic advice, the terms of this amendment should not be hard to fulfil. Indeed, some might see it as a necessary benchmark to ensure the accuracy of that information and a crucial step in reassuring the public and the media of the worth of private pension saving. I am particularly worried about the activities of the media in this area.

This amendment does not seek to drive the Government in a different direction from where they say they intend to go. Instead, it seeks to ensure that decisions around the implementation of auto-enrolment and personal accounts will continue to take account of this issue, even after the Bill has left Parliament. I beg to move.

I wonder whether the noble Lord can help me. I am not sure that I fully understand the intent of proposed new subsection (2), which states:

“If the projected figure published in accordance with subsection (1) exceeds 10% of the total expected pensioner population”.

Does that mean 10 per cent of the pensioner population, or 10 per cent beyond the pensioner population expected to be on benefit, given that about 25 per cent of pensioners draw housing benefit now—that figure is not likely to fall—and that some 30 per cent, perhaps more, will continue to draw council tax benefit? Is he saying that all this must come into play given 10 per cent of a retiring population, or the full existing pensioner population, or is the 10 per cent greater than the pensioner population already in receipt of benefit? Which of the three possible meanings is intended under the proposed new subsection (2)? That may help us to further the debate.

Proposed new subsection (2) means much as it says. When the number of people on expected means-tested benefits exceeds 10 per cent of the pensioner population—I put the word “expected” in another part of the amendment; I hope that makes it clearer for the noble Baroness—a report will be published and the introduction will be delayed.

Perhaps the noble Lord could help me further. If in 2011, say, the Government were to publish a report stating that they expected that of the total pensioner population in 2012, instead of 30 per cent there might be 25 per cent on council tax benefit, and that instead of 25 per cent on housing benefit, as now, there might be 22 per cent, does that mean that, according to the rest of the proposed new subsection (2), the whole of this Act must be postponed?

Can the noble Lord suggest how, under any circumstances, he will remove council tax benefit and housing benefit from an existing pensioner population—all 11 million or so of them—who are in their existing housing with their existing income and entitlement? How will he bring the number down from 30 per cent or 25 per cent to 10 per cent in order for the second part of proposed new subsection (2) to come into play without making them poorer? Does he propose that they should have that benefit withdrawn, or that the local authority should cut rents, or that the Government should halve council tax? Given that he is dealing with the existing pensioner population, he cannot have more than marginal changes at the edges of the statistics.

The noble Lord is therefore saying that the amendment would in effect delay the implementation of the legislation to the disarray and discomfort of all the stakeholders involved who have been planning to this timetable. There is no way in heaven or hell that you can deliver proposed subsection (2), given the existing pensioner population who are already in the council houses and flats and housing association homes and claiming council tax benefits now. The amendment is nonsense. It is simply a recipe for delay that would disrupt the entire planning process for all the stakeholders who are working to the 2012 deadline. I suggest that the noble Lord thinks again.

I shall be delighted if the debate on the amendment means again that not only two can play. I agree with what I am sure is the intention behind the noble Lord’s amendment—to highlight the continuing effect and amount of means testing in the system and to see what can be done about reducing it or at least ensuring that the most high-risk groups are properly protected from it.

We have already discussed the basic statistics and the basic problem at some length on the amendments that my noble friend Lady Thomas and I moved on a previous day to try to ensure that generic face-to-face advice was available to the high-risk groups, so I do not want to go over that again. I should say to the noble Lord, Lord Skelmersdale, that the DWP recently published its estimates—I do not expect that those figures will change very much—of the proportion of pensioner households that will be subject to means testing at different dates. They are, of course, exceptionally high. I remind noble Lords that under the new reform system, as the department calls it, the figure for 2020 is now 55 per cent and still 40 per cent for 2050.

It might be worth pointing out that the DWP says that it is testing the sensitivities of these assumptions. It states that these four figures,

“have been tested for sensitivity to plausible changes in underlying assumptions about the future level and distribution of pensioner incomes and future growth in rents and council tax liability”,

which is obviously significant. It continues:

“Based on these tests the proportion of pensioner households entitled to income related”—

in other words, means-tested—

“benefits in 2050 appear to be robust to the changes in assumptions, with pension credit varying by around five percentage points”,

and the other benefits by very little.

We all know that there will be a serious problem with means testing for the foreseeable future. Although, if pressed, we would support amendments that seek to ensure that the Government update those estimates, in a way this amendment does not go far enough. I hope that all of us who are concerned about this will think carefully about this before Report. We need not only to know the shape of the figures, which are fairly clear, but to press the Government further on what we are going to do to protect the most at-risk groups.

I add to what my noble friend has said by discussing how auto-enrolment into pension provision will interact with those who are likely to be on means-tested benefits in retirement. This is one of the chief concerns of all those who are involved in the Bill, and is one of the great unknowns. The noble Baroness, Lady Noakes, said at Second Reading that it was,

“something of an evidence-free zone”.—[Official Report, 3/6/08; col. 119.]

Even the redoubtable Pensions Policy Institute said that some of the factors that will have an impact on returns from savings and personal accounts,

“could be more problematic than others”,

in predicting the point at which saving decisions are taken. It continued:

“Clearly, nobody can predict with certainty all of their future life circumstances”,

or marital status, and it may be difficult for an individual to predict whether he or she will be renting in retirement. Although it is clear that more pensioners will own their own homes in future, no one can know for certain what will happen to the housing market. Those who may have thought that their homes and jobs were safe may find that things are very different some years hence.

In his winding-up speech at Second Reading, the Minister did not give us very much information about the Government’s review of Pays to Save. He just said that the scheme was not buying people off, that it was there for a proper purpose and that he was sure it would deliver. Perhaps in his reply he could expand on what action the Government will take and in what circumstances when the review is published.

From the very beginning there has been concern on all sides about the interaction of the scheme with the benefits system. Therefore, I support the intentions behind the amendment but, on the other hand, it does not meet the objectives which I think the noble Lord, Lord Skelmersdale, had in mind in introducing it. In particular, I share the views expressed about proposed new subsection (2) of the amendment. I do not see why the implementation of the scheme under Clause 58 should be postponed for at least 12 months, as the subsection says, while consideration is given to the projected figure of 10 per cent referred to in the amendment. That seems a rather unusual figure to quote. In those circumstances, although I am sure that the Committee is concerned to ensure that interaction with the benefits system should be ironed out somehow or another—indeed, we know that the Government have that under continuing review—I cannot support the current wording of the amendment.

I ask the Committee to reject the amendment and I thank all noble Lords who have contributed to this short but important debate. As the noble Lord, Lord Oakeshott, said, the issue of savings incentives and the interaction with means-tested benefit has enjoyed lengthy and constructive debate in both Houses, and it is clearly an important issue. Perhaps he did not use the word “constructive”; that is my gloss on his presentation.

I reiterate that under reasonable assumptions the majority of those who enrol can expect to benefit from having saved, and we must not lose sight of that. The noble Lord’s amendment would delay introduction of the personal accounts scheme, but seemingly not auto-enrolment generally, if the number who may not benefit was greater than 10 per cent of the total expected pensioner population.

I hope that we all accept—the noble Baroness, Lady Thomas, touched on this—that no amount of analysis can tell individuals at the point of enrolment exactly how their lives will turn out possibly decades into the future. We know from analysis, for example, that there is around a 10 per cent chance that a 20 year-old man today will not reach state pension age, but no one is suggesting that we should stop advising 20 year-old men to save for their retirement. On this issue, perhaps I may quote the noble Lord, Lord Turner, who said that,

“if at the end of the year your house has not been burgled, it does not mean that it was bad advice to buy a household insurance that year”.

As I said at the outset, our package of reforms means that most people can expect good outcomes from saving, while those who do not wish to save will be able to opt out. The provisions in the Bill represent the remaining pieces of our pension reform package, and any delay in their introduction would deny many individuals the very real benefits of starting to save early.

Again, as the noble Lord, Lord Oakeshott, said, we have already published our projections of entitlement to all income-related benefits for the pensioner population until 2050. Around 40 per cent of pensioner households will be entitled to one or more income-related benefits by 2050. As my noble friend Lady Hollis said, there is no prospect of reducing that figure to 10 per cent. Indeed, as I think we have discussed before, if the basic state pension were doubled in 2012, in 2050 25 per cent of pensioners would still be on one or more income-related benefits.

Through our reforms we are reducing the level of means-testing, not increasing it, because without reform it would have increased to 75 per cent in that period. By 2050, the basic state pension will be worth more than double what it otherwise would have been, ensuring a solid foundation for private saving.

We could debate, as we have done in the past, what an appropriate level might be for pensioner households on means-tested benefits, but I do not apologise for the need to retain a means-testing system. Such a system protects the most vulnerable individuals against poverty within inevitable cost constraints. However, in this debate we must absolutely not lose sight of the fact that both our analysis and that of the Pensions Policy Institute demonstrate that people can be on benefits in retirement and be better off for having saved. Equating those on income-related benefits with those who will not be better off from savings is simply not right. We need to remember that because of our reforms to the state pension system, someone with a good record of working or caring, including those on low earnings, can expect a state pension of around £150 in 2008-09 earnings’ terms by 2050, more than £25 above the pension credit guarantee.

The interaction between means-testing and savings incentives is not a new issue. There are already safeguards in place to improve incentives to save. In 2003, we introduced the savings credit to prevent a pound-for-pound clawback. People can take up to 25 per cent of their pension pot as a tax-free lump sum and those with less than £16,000 can take 100 per cent of their pension pot. Our analysis shows how these are already working to improve incentives for those who have been suggested as the most vulnerable. Older groups with lower earnings, for example, are most likely to be able to trivially commute their savings, gaining a cash sum. Up to £6,000 of that—a significant amount for many, I suggest—can be held without any impact on any pension credit or housing benefit that they may receive.

We would have to think carefully before discouraging such people from saving. Indeed, while the challenges around means-tested benefits are clearly an important matter, the PPI has stated that this,

“does not mean that people should … be auto-enrolled”.

But, as the PPI has also made clear, this implies that people will need access to clear information. We fully agree with that; indeed, Clause 9 of the Bill recognises the importance of information. We will ensure that every automatically enrolled individual has access to the range of information that they need to understand the process of automatic enrolment, the pension scheme into which they will be enrolled, their expectations for a state pension and the implications for their later life.

Anyone who does not think that pension saving is for them, regardless of the reason, will be able to opt out. But that should not be allowed to detract from, or hinder, the fact that we are further improving savings incentives through this Bill. Millions of workers, many for the first time, will see their pension contributions matched pound for pound through the employer contribution and tax relief. It is right that we should protect against unpredictable events and provide a safety net for those who need it most and it is right that we should encourage individuals to save and take personal responsibility for their own retirement income, giving them the security and flexibility of their own savings pot.

Governments always need to strike a balance between alleviating poverty and encouraging savings. They must do so in a way that is both affordable and sustainable. Our reform package achieves that. We agree that this is an important matter, but the Government, external stakeholders and, I thought, all political parties are also agreed that there is no magic bullet and that progress of this Bill need not be delayed. Nigel Waterson MP said:

“We never thought it was realistic to deal with this during the life of the bill”.

That said, we have listened carefully to the concerns of those in Parliament and outside. Age Concern, Help the Aged and others have asked us to set up a review to look at these issues again and to have an informed debate on the issue. We are now well advanced in that review involving stakeholders and external experts as well as the very best analytical minds and tools at the Department for Work and Pensions. The review will further expand an already extensive evidence base on this issue. It will further improve our understanding of incentives and the impact on behaviour. It will look at how different factors and individual characteristics affect outcomes. Clearly the lessons of this will be vital in developing our information strategy. Importantly, it will assess some of the proposed measures put forward, putting in the public domain an objective assessment of their pros and cons, their costs and the effects on our wider tax and benefit policy.

In all this, we are mindful of the need to strike the right balance between alleviating poverty and personal responsibility for retirement income. We are also mindful of the cost constraints that we face, but we will consider the evidence of the programme and its full implications. I shall briefly touch on some technical issues relating to the drafting—

Before the Minister does so, perhaps I could say that I was rather surprised to hear what he said in his first reference to the Pensions Policy Institute. He gave the impression that the PPI—I know that it guards its independence very jealously—was in some way endorsing government policy. Although I do not ask him to read it out again now, has he cleared what he said with the PPI and, if not, could he do so please?

As it happens, I met the PPI last week; I discussed the programme and the work that it had done and sought its views. The noble Lord is right that it is an independent body, but it was a clear statement; I think that it was in part of the PPI’s response to the consultation and in its report, which indicated that the challenges around identifying those at risk should not cause people not to be auto-enrolled. It meant that people needed very good information, and that is where we agree.

I asked whether the Minister could send the sentence that he used to the PPI and make sure that it is a fair reflection of its views. That is all that I am asking.

I am happy to do that, but I asked a question when we met and I believe that what I said was a direct quote from the letter that I was sent. However, I will certainly clarify this with the PPI if it helps the noble Lord.

Let me make clear some of the technical drafting. My noble friend Lady Hollis dealt with people being at risk, which is not defined here. It seems that the noble Lord is equating the numbers on means-tested benefit with those who are at risk. On that basis, the 10 per cent threshold could not be met.

The amendment would prevent the development of personal accounts, which is not the same as auto-enrolment. It would permit auto-enrolment to proceed on the basis of the noble Lord’s formulation but not the development of personal accounts. That cannot be right.

There are some other technical issues, but I do not want to dwell on those, as we are focused on a more substantive debate today. Let us not forget the big prize here. This package will benefit millions of people. We estimate that in 2050 the reforms will lead to a total increase in annual private pensions income of £11 billion to £16 billion at today’s prices. That is a prize worth having.

I am advised that I read out the PPI quote wrongly, missing out a “not”. I will now read it out properly. I said that while this is clearly an important matter, the PPI has previously stated that this,

“does not mean that people should not be auto-enrolled”.

I thought that that was what I had said but, if not, I am happy to clarify it.

Even if we were to define and identify accurately an 11 per cent risk group, as the noble Lord wants, there would be justifiable fury if we denied the benefits of auto-enrolment to 89 per cent of people who were not at risk, so that millions of low to medium earners would not get the chance of having what many of us have taken for granted throughout our working life. I hope that, on reflection, the noble Lord will feel able to withdraw the amendment.

I had no intention whatever of pressing the amendment. I accept straight away the criticism of the noble Baroness, Lady Hollis. I accept, too, that I went way over the top in drafting the amendment, although I hope that I did not go way over the top in speaking to it. As the noble Baroness, Lady Turner, and noble Lords from the Liberal Benches have said, this is an extremely important subject. There is no doubt in anybody’s mind that means-testing is essential to the success of what we all want to see: a significant increase in the number of people saving for their retirement.

It is inevitable that jobholders will exercise their right to opt out if they feel—and it may not be a particularly sustained feeling—that investing in a personal scheme, especially towards the end of their working life, would so eat into their state benefits, even eliminating them altogether, that they would be no better off than if they had not saved into such a scheme. The advice that they are given, and that we have discussed already, must—simply must—tell them whether it pays for them to save in this way.

I referred in my opening remarks to the work that the Government are still undertaking on this, which the Minister called a review. This, as I understand it, will be completed during the Summer Recess—I hope that I am right. It is on the basis of those results, or what I can discover about the review if it is still going on, that I will decide whether to come back to this matter at a future stage of the Bill, because I am absolutely convinced that it is essential to the future of the whole scheme. I beg leave to withdraw the amendment.

Amendment, by leave, withdrawn.

I beg to move that the House do now resume.

Moved accordingly, and, on Question, Motion agreed to.

House resumed.