Work and Pensions Committee
Oral evidence: Protecting pension savers – five years on from the pension freedoms: Saving for later life, HC 989
Wednesday 23 February 2022
Ordered by the House of Commons to be published on 23 February 2022.
Members present: Stephen Timms (Chair); Debbie Abrahams; Siobhan Baillie; Selaine Saxby; Dr Ben Spencer; Chris Stephens; Sir Desmond Swayne.
Questions 1 - 40
Witnesses
I: Chris Curry, Director, Pensions Policy Institute; Carl Emmerson, Deputy Director, Institute for Fiscal Studies; Nigel Peaple, Director of Policy and Advocacy, Pension and Lifetime Savings Association; and Rob Yuille, Head of long-term savings policy, Association of British Insurers.
Written evidence from witnesses:
Institute for Fiscal Studies (IFS)
Association of British Insurers (ABI)
Pensions and Lifetime Savings Association (PLSA)
Witnesses: Chris Curry, Carl Emmerson, Nigel Peaple and Rob Yuille.
Q1 Chair: Welcome everybody to the first evidence session of the Select Committee on Work and Pensions inquiry on saving for later life. We are very grateful to the four expert witnesses who join us this morning—thank you all very much for being with us. May I ask each of you to say very briefly who you are?
Chris Curry: Good morning. My name is Chris Curry. I am the director of the Pensions Policy Institute.
Carl Emmerson: Good morning. My name is Carl Emmerson. I am deputy director at the Institute for Fiscal Studies.
Nigel Peaple: I am Nigel Peaple, director of policy and advocacy at the Pensions and Lifetime Savings Association.
Rob Yuille: I am Rob Yuille, head of long-term savings policy at the Association of British Insurers.
Q2 Chair: Thank you all very much for being with us. Our current pensions system—a flat rate state pension linked to earnings plus workplace auto-enrolment—was recommended 16 years ago by the Turner commission when I was Pensions Minister. Is it still right, or does it need now to be changed in some respects?
Chris Curry: As you say, that was quite a long time ago. I can still remember the discussions that happened then. I think what is interesting is how the commission formed a consensus around that particular solution at the time, how well that consensus has remained up until now, and how it remains to this day. From my perspective, what is interesting is that the solution still forms the basis of the system as we can see it going forward, and lots of things about it work well.
What is different, and what has changed—a lot has obviously changed in the last 16 years—is the environment in which that system is operating, and what has happened outside the pensions system, as well as some significant policy changes within the system. I think we may potentially have lost sight of what the objectives of the pensions system actually are.
I think that, although the recommendations of the Turner commission and the analysis behind them were almost exemplary in the way they were formed, and they still hold true today, we need another examination to try to test exactly what we think that system is trying to achieve. Only by doing that can we then determine whether the system is actually achieving its aim. As a way of trying to help that, we at the Pensions Policy Institute have been working with Aviva on a pensions framework, which we published just before Christmas. That looks at numbers of key indicators and metrics of measurement around three important principles: adequacy, which is obviously a very important topic for today, sustainability, and fairness. It is only by looking at those very broad issues and trying to determine what we think the system is trying to deliver that we will be able to determine whether the current system can deliver against those objectives.
Q3 Chair: Do you think there needs to be a new commission to ask those fundamental questions, or is some other approach needed?
Chris Curry: I think it would be difficult to have a new commission. What happened back in 2003 to 2005 was almost—in a positive way—like the perfect storm: we had some very, very good commissioners; they had a lot of support and analysis; it was the right time in the system’s history for there to be one of those important overviews. I think there is still a need for further consensus on how we take this forward, but I do not think we are in the same kind of fragmented situation that we were in back then. As I said, there is a fair degree of consensus that automatic enrolment is the right way forward, and that the state pension providing a firm underpin is the right way forward. I think there are issues to discuss around what happens on top of automatic enrolment, the size of automatic enrolment and how we do it, but the framework is actually pretty secure.
Carl Emmerson: I certainly think that stability and consensus are attractive features of a pensions system where you are asking individuals to make long-run decisions, and that many of the things the Pensions Commission recommended, which, as Chris said, have been implemented during successive Governments, of different colours, were very, very welcome. But I also would stress how much has changed since then. The economic environment is so different. We have had a financial crisis. We have had a pandemic. Interest rates are at levels that are so low and are expected to stay very, very low for a long while. We have seen house prices grow and home ownership rates fall sharply. We have seen changes in the labour market, with growth in the gig economy and growth in self-employment. We have also seen radical policy changes that the Pensions Commission did not envisage, such as pension freedoms and lifetime ISAs. So much has changed that I think this is a moment to take stock and have a review. I wouldn’t be surprised if something like the state pension we currently have is a good underpin, going forward. I wouldn’t be surprised if there are lots of lessons from the automatic enrolment reforms and the state pension age reforms that we want to build on. But I also wouldn’t be surprised if there are other things we need to look at and think about that the commission just couldn’t have predicted at that time.
Nigel Peaple: By international standards, the UK state pension is at the modest end of things, but it does seem to provide quite a good underpin to head off poverty. When you look at various measures such as the Joseph Rowntree Foundation minimum income standards, which look at what people need and what people think is necessary, you find that the state pension is about right; it’s maybe about £1,000 less. I will just mention at this point that, as you are aware, our Retirement Living Standards are basically on the same methodology and the same as the JRF’s at that minimum level. So the basic state pension is probably about right; it is a bit on the lean side.
I think that workplace pension saving really is too low. To look back at what the Pensions Commission did, they looked at the question of what the target should be and they said, “Oh, there’s lots of contradictory information on this; it’s quite hard to make a judgment,” but they came up with a solution whereby they said someone on median earnings should probably aim for about two thirds of former income; someone on half of average earnings probably needs about 80% of former income; and someone higher up the scale—say, on twice average earnings—probably needs only about half as much as they were on previously. So the commission had these targets.
In 2016 and subsequently, we have done modelling with independent actuary firms to say, “How much money do people have now? What saving are they doing? Will they get those target amounts that the Pensions Commission talked about?” And the conclusion from that modelling is that they won’t. Only about half of people are likely to get those amounts, and they are the people with some or current DB pension provision, which, as you know, tends to involve much higher contribution levels. Of the people who are in the defined contribution saving category, only 3% are going to hit those Pensions Commission targets.
Therefore we think that actually it is very important to up the contribution levels of automatic enrolment. The Pensions Commission originally said, “We think people need about 16% of a band of earnings, to get to these targets, and we think the right thing is to go for about 8% through automatic enrolment, so that people don’t have to think about it, and 8% through voluntary saving.” But what we have seen is that the voluntary saving side hasn’t come through. What we have seen is the behavioural economics lessons of automatic enrolment. This works really well, with 90% of people staying in. They are very happy about it. They can opt out if they want, but they like it. So we believe the answer is to be increasing the automatic enrolment contributions and, as you may have seen, our view is that you could go from the current 8% of band earnings to about 12% of all earnings over the next decade—there’s no need to rush it. Very importantly, given the cost of living crisis at the moment, it shouldn’t be rushed. Ideally, you might do this in two steps, the first step being to increase the employer contribution from 3% to 5%, which would match the employee contribution. If you did that by 2030, it would be a levelling up of pension provision, and then you could try to add on an extra per cent. or so after that. So we think that workplace pension saving really needs a bit more going in. There are many issues around this, which I know we will explore later.
If I may, I will add one more thing, just on voluntary private pension saving. There isn’t really that much of this about, and this, again, is because people find it really hard to think about the long term and really hard to put money away for the future, rather than meeting the very many pressing needs that are on them. Things are being done at the moment to try to make it easier for people to engage with things. You have the pensions dashboards programme, and you have things like our initiative on Retirement Living Standards to bring that to life. But frankly, if I were a betting man, I would put my money on the automatic enrolment lessons rather than—well, let’s try the engagement path, but I wouldn’t really be confident it will come through.
Rob Yuille: I agree with much of what has been said—not everything. Your question was, “Is the current system still right?” Yes. Does it need to change in some ways? Also, yes. I know that we always say it, but it’s worth repeating that automatic enrolment has been phenomenally successful in increasing workplace pensions participation, but we do need to see contribution levels increase. How do we go about that? One of the benefits of the Pensions Commission, as it was, is in having a long-term strategy, as well as a consensus-driven approach and cross-party support.
We now need to look more holistically at issues beyond pensions, such as debt, housing—as Carl mentioned—and non-pension savings. If that is to be achieved through the Pensions Commission, that’s great, but we really need a long-term strategy. To pick up on the point that Nigel mentioned, it is not an either/or between voluntary savings and increasing auto-enrolment contributions. There is a great deal of voluntary saving, but that clearly isn’t done by everybody—I can come back to that later if we talk about the self-employed, for example. So we should be looking at both of those things.
Q4 Selaine Saxby: I am looking at assessing adequacy. Obviously, there are two main approaches to that assessment: the retirement income standards—where you have your cost of living based on a basket—and the replacement rates, where we are looking at income in working life. What are your views on the best approach?
Chris Curry: I think they both have their merits, and I think that which one is the best depends very much on what you are looking at and whose perspective you are looking from. They measure different things.
The minimum income standards and the approach taken with the PLSA Retirement Living Standards is something that probably resonates more with individuals, so it is something that they can understand, and see what it is they might be able to achieve. Whether they can understand what that will actually mean for them in retirement, and how that will change during retirement, is another question, but it is certainly useful in trying to help people see what it is they might want to try to achieve.
However, if you are thinking about it in policy terms, that may or may not be the most useful way of doing it, and the advantage of the replacement rate approach that the Pensions Commission put forward was that it kind of helps maintain a standard of living for individuals, so keeps them almost in the manner to which they have become accustomed when they get into retirement from their working life. It will not be exactly the same, but it will not be a massive drop in standards and something that they may come to regret when they get there.
Rather than the definition, I think there is a wider question around what we actually want to include when we are talking about adequacy. I think it has already come up a couple of times today that, on the first instance, the pensions system is now very different, and many future pensions will not be an income stream but a capital lump sum, which people withdraw or take from their defined contribution pension scheme. In those kinds of circumstances, having an income target doesn’t necessarily match up with their experience or what they’re planning to do, so is there a case for including some kind of capital measure within the adequacy target as well as an income target? That is one aspect.
The other aspect, which, again, has come up through things like freedom of choice and voluntary saving, is that a lot of voluntary saving for retirement is no longer done through the pensions system. So it is important that we take account of what else is happening outside the pensions system to determine what assets people have to live in retirement.
To emphasise that, as Nigel mentioned earlier, some of the analysis that we, and others, have done suggests that of the people aged 50 and above, but before retirement, in the UK, about 45% of them are not likely to end up meeting a target replacement rate, based purely on their pension income. However, if you include other assets as part of that, that actually falls to 35%. If you take account of some of their housing assets too, then it falls even further, to below 30%.
So if you bring in other assets, rather than just pensions, you can change the numbers of people who you think might be comfortable in retirement in different ways. Obviously, there are challenges in that. How do you measure those things? How do you assume that, just because they have an asset now, it’s going to last until retirement? We also need to take account of other things, like debt. However, I think there is an opportunity to think much more broadly about what retirement actually means now, and how it’s supported, and to try to broaden those definitions of adequacy to incorporate other sources of income and assets too.
Carl Emmerson: I definitely think there is a role for both measures. On the adequate income measure, one thing I would stress is that while the state pension might not be a fantastic level of income for many, it is much more generous than what we provide to working-age individuals, and the gap between the two has been growing.
If you go back 30 years and look at the income we would give a single person not in paid work, ignoring their housing costs, and if they are just below the state pension age compared with just above, the state would give them about 30% more if they were just above the state pension age. Now, however, we give them about 130% more, so that gap has just grown and grown. When thinking about poverty and who has adequate levels of resources, we need to think about that big cliff edge. It makes it really important that you get the state pension age in the right place for many people.
On replacement rates, clearly that is what most people are going to be focused on, as Chris said. They are going to want a living standard in retirement that is in some way comparable to what they got to enjoy through their working life. The Pensions Commission approach was to ask people about what kinds of replacement rates they would like, and that is clearly one way of going about it.
Some colleagues at the IFS have constructed a simple economic model to ask the question, “Under certain assumptions, what replacement rates would an economic agent want?” That sheds light on what kind of factors matter, and it is really quite interesting. If the rates of return are lower, the model suggests that people want lower replacement rates. If we buy that, we should not be surprised if, for example, people in today’s world want lower replacement rates than people might have wanted, say, 20 or 30 years ago, when rates of return were much, much higher. Essentially, to get a certain level of retirement income, you have to give up more today, because the price is higher. That is what lower interest rates mean.
We also find that many other factors affect what replacement rates people might want, such as how many children they are going to have. If you are going to have more children, you will want a lower replacement rate than if you are going to have fewer children. Another is how much uncertainty there is in your life about your future earnings. If there is more uncertainty, you will want to put more away, and that means, on average, you will achieve a higher replacement rate than people who are perhaps a bit more sure about how much they will earn. I think I would want to stress that these replacement rates that we tend to construct and use in policy terms are very broad brush and will not apply to everybody.
Selaine Saxby: Thank you. Nigel, anything to add?
Nigel Peaple: I agree with Chris that target replacement rates are very useful as a policy tool when designing the system. I also agree with Chris that the Retirement Living Standards has a tremendous capacity to engage individuals because they can really picture their future and understand all the different facets of lifestyle, giving them something to aspire to. The first of those is effectively a kind of expert-led view, albeit based on research—not sure where I’m going with this. The second of those, Retirement Living Standards, is based on extensive research with people from Britain across the whole UK, asking them, “What do you think works in our society as a minimum level? What do you think is a comfortable level? What is a moderate level?”
What I find quite interesting is that, actually, the two figures come out in quite a similar place, with 10, 20 or 30. Take the example of the median, where there is this old established sense of, “Oh, you go for about two thirds of income.” That’s what the Pensions Commission said and what people told them. Median full-time earnings are about £29,000 or £30,000 a year, and therefore that might lead you to think that they would want an overall income, including the state pension, of about £20,000 a year. When you look at the Retirement Living Standards research, it comes out in the early £20,000s, and you get a similar alignment at the minimum level and at the higher level—the 30k level. I am wondering whether there may be some sort of feedback loop about reflecting the society we are in and people’s sense of what average earnings are and what their aspirations are.
What I really would like to say is that I do feel that this is one of the most important aspects of the pensions system: having a clear target that everybody can then work towards, whether individual providers and advisers or savers or the Government.
The only other thing I would like to say on this, just picking up on the assets side of things—it is quite brave of me to take on Chris on an assets issue given that he is the PPI—is that when I last looked at the Wealth And Assets Survey I was struck by the fact that people on median earnings do not have a lot of wealth. They do not have very much in the way of accessible financial savings. They do not even have that much in terms of property saving. They have something more for pension saving. Although some people, it is true—I am sure Chris will probably agree with this—will be able to draw on property wealth, often we are talking about people in the upper elements of the income distribution, meaning people in the top 25% of earners in society. I think we need to bear that in mind when thinking about what the UK regime is and what works for the average person.
Rob Yuille: I do not need to add too much more to that, but, from a consumer perspective, policy makers need all those measures and more. Critically, they need them where there are gaps and inequalities in adequacy between different groups. I am sure we will come on to that.
From a consumer perspective, it is a lot to get your head around. A pounds figure will be more relatable; we know that levels of financial capability and numeracy in the UK are low. Yesterday we announced a partnership with Plain Numbers that will help providers to communicate in a way that people can more easily comprehend. It also illustrates the importance of dashboards, so that people can see all their pensions together, and how they are progressing towards a target that they might set, or that might be set for them. They may potentially see that alongside their other wealth. It is worth bearing in mind that a lot of people will assume that they are already on track because they are saving at a level that has been set by the Government.
Q5 Sir Desmond Swayne: Chris, of the 11 million people between the age of 50 and retirement age, apparently 3 million are not even going to secure a minimum income. We are not saving enough, and we are facing a crisis of pensioner poverty—is that correct? Are we sufficiently aware of it? Why is it so much more difficult for the younger generations?
Chris Curry: There are a number of different things in that question that I think are worth trying to unpick. The figure of 3 million people that you cite is definitely correct; it is from research that we did looking at the numbers of people likely to hit different income targets in that group.
You are right that it is quite hard to envisage someone who is not able to hit that very low minimum income standard when the state pension is relatively close to that, as Nigel said earlier. The fact of the matter is that we still do not have completely full coverage of the state pension; even with the new state pension, not everybody receives the full amount—there are gaps. There are people who, for whatever reason, have not benefited from the introduction of automatic enrolment, especially in that older age group. People such as those aged 50 to 65, who may not now be part of the labour market or haven’t been since the introduction of automatic enrolment and so have nothing else to rely on.
It is difficult to say whether there is a pensioner poverty crisis. The numbers for pensioner poverty have been steadily coming down, although recently they have started to tick back up again. I think there are just as many questions about pensioner poverty for those already over retirement age—who may be above poverty when they enter retirement but potentially slip into poverty as they go through retirement—as there are for those who are coming up to it.
It is very difficult to see what other measures could be taken for that group of 50 to 65-year-olds that would substantially help them, unless it is direct intervention through the state. It is unlikely to work through the labour market. That is even more the case if we look at recent figures that suggest that increasing numbers of people are leaving the labour market aged between 60 and 66 as a result of the pandemic that we are still going through. It is a difficult question to answer. I do not think that it is unknown that we have these issues; I think it is very difficult to determine exactly what to do about them.
For the younger generation, I think the difficulties are harder to untangle. The pension system is probably better set up for them; I think they can expect a better state pension underpin, and having spent longer under automatic enrolment, better DC pensions than 50 to 65-year-olds. Obviously, they are not going to benefit to the same extent from defined-benefit pensions, which is quite an important aspect, although those in the public sector will still have access.
What is more likely to affect them is the uncertainty in the labour market and the stagnation in earnings, which means that they are not seeing the same progression that previous generations have seen. That means that at the same stage as previous generations, they are on lower earnings and further behind in saving to get on the housing ladder. Despite automatic enrolment, they may not see pensions as a priority, so there is not that voluntary saving expected in order to get them up to the replacement rates and adequacy targets that we have talked about.
Young people do still have time—I think that is the difference. Although they may appear to be in a worse position now, there is still an opportunity between now and the time that they get to state-pension age, or to the end of their labour market experience, to either increase their engagement and use of the system or to evolve the system to take account of changes we are seeing in the underlying economy.
Q6 Sir Desmond Swayne: You have identified groups that are specifically at risk—low earners, women, minority ethnic groups, the disabled and the self-employed. Do we know anything like enough about the discrete problems within each of those definitions to be able to see what the drivers are?
Chris Curry: We can see the first-order impacts, but it is difficult to understand what is driving those. In reality, all those groups suffer from their labour market engagement. The UK pension system is very heavily dependent on what happens to individuals during the labour market, especially when you get into defined-contribution pensions. The amount of the time you work, and the amount of the earnings that you have, has a very strong impact on the amount of pension that you receive if you do it through a defined-contribution pension system. It has been improving in recent years because of the improvements in the underlying state pension.
The new state pension has actually been very good at improving state pension outcomes for those under-pensioned groups, as we call them, but it has not really had an impact on their private pension saving. As we were saying, the whole outcome from the UK state pension system is a combination of the state plus workplace pension plus voluntary saving. For those under-pensioned groups, workplace pension does not work as well, even in automatic enrolment, because they are either lower earning or intermittent earners, so they do not benefit in the same way from consistent contributions from employers, consistent tax relief and consistent own contributions in that respect. Because they are lower earners with lower income, they are very unlikely to be in the voluntary saving market either.
What can be done is a lot around the labour market, and I think what is being done is a lot around the labour market and looking at participation. The difficulty is trying to understand what it is that you can do in the labour market that will increase participation and increase earnings in these particular groups.
Q7 Sir Desmond Swayne: Carl, the IFS believes that some people are saving too much and, as a consequence, cannot live adequately now. How do we get that balance right? How does the Government achieve that?
Carl Emmerson: The research you are referring to has been looking at automatic enrolment and some of its effects, and I certainly agree with what was said earlier about that being a big, successful reform that has achieved very high levels of coverage. We have looked a bit more into some of the details, though. Prior to automatic enrolment, higher earners were much more likely to be in a pension than lower earners, and older people were much more likely to be in a pension than younger people. Automatic enrolment essentially levels up a lot, so you end up with coverage rates of 85% or 90%, but that is pretty much true across all groups in society. You are getting bigger effects among young people and bigger effects among lower earners.
Then we took a group of people who look like they are in quite a bit of hardship at the moment. You can take people who are in arrears on their council tax and you can say, “What was their pension coverage prior to automatic enrolment?” Some were in a pension, but it was a minority. You can say, “How many are in a pension after automatic enrolment?”, and you find that 85% or 90% of them are in a pension. If they opted out, I don’t know whether they would choose to use the resources that they had available at that point to pay off their council tax bill or whether they would spend it on something else, but I certainly question whether coverage rates should ideally be that high among groups like that.
I am not saying you should get rid of automatic enrolment at all but, at the margins, if there are ways of making opting out easier for some groups that did not imperil the high coverage rates among others, I think that would be a good thing.
Q8 Sir Desmond Swayne: Rob, the ambition is to reduce the age for auto-enrolment from 22 to 18, and I understand that you believe that should be phased in over a period of two to four years by the mid-2020s. Does anyone think it should be sooner?
Rob Yuille: I will pick up briefly on Carl’s point as well. One would hope that those people would not be in council tax arrears forever, so it is not a question of whether they are in or out of pension savings. It is a question about transition and making it as easy as possible for people to cease their own contributions and then re-enrol. That is the first point.
The automatic enrolment review recommendation is by the mid-2020s. There is a very broad consensus that they need to be implemented, and soon. Yes, they should be phased in. Arguably, that phasing in has already started, because the lower qualifying level of thresholds was frozen this year. It does not primary legislation to reduce it, so that could potentially happen alongside the changes to net pay in 2024-25.
The reduction from 22 to 18—the age threshold—does need primary legislation. That might take a little longer, but yes, it should be phased in. The history of auto-enrolment is doing things carefully and in stages, so as not to disrupt the consensus and the success that we have had. There are also important reasons for doing it in that sequence, so that the small pot problem that we have is not exacerbated, so that younger people who are more likely to earn just over the earnings trigger and who contribute very small amounts above the level of qualifying earnings would generate a small pot. So, yes, it should be phased in, and we should start soon.
Q9 Sir Desmond Swayne: Is there general agreement on that?
Nigel Peaple: Yes, taking account of the current cost of living pressures, but certainly middle 2020s to be introducing these things in a phased approach.
Carl Emmerson: I agree, although I question why we need age ranges at all. I do not know why we would want to exclude 17-year-olds or 67-year-olds from automatic enrolment and the ability to get an employer contribution. I question why we would have them in the long run.
Q10 Sir Desmond Swayne: In answer to an earlier question, Nigel, you said that there should be an immediate increase in the employer contribution—
Nigel Peaple indicated dissent.
Sir Desmond Swayne: Am I wrong?
Nigel Peaple: Not immediate.
Sir Desmond Swayne: By the end of the ’20s—or was it the early ’30s—up to 12%, and thereafter increasing.
Nigel Peaple: Very nearly that. I suggested that, ideally, we get the employer contribution up in the late 2020s, by about 2030. So, up by 2% in the late 2020s, and then in the early 2030s to go for the 12%, with 1% extra for the employer and 1% for the employee.
May I make one quick point? Historically, in the UK, workplace pensions were two thirds employer and one third employee. Under automatic enrolment, that has been reversed. We are arguing—and many people in the industry feel—that it would be better to get to at least 50:50 and to level up the employer contribution.
Q11 Sir Desmond Swayne: Is that going to be enough? Is there a danger that people will opt out, because it is actually too much? Ultimately, for those people who are approaching retirement, those gradual increases are not going to do very much. What can we do for them?
Nigel Peaple: Is it enough? These figures are based on the assumption that the Pensions Commission suggestions—two thirds for the median, half for someone on a higher income and 80% for lower income—are about right. Given what we see from our retirement living standards research on people across the UK, that feels about right—it is really important to have that target—so I think that 12% probably is about right.
Is there a danger of higher opt-outs? At the moment, as you know, the level of people staying in is at about 90%. People are very happy. As has been mentioned, people assume that Government have sorted their pension for them, and they are up for it and they stay in; if they are in difficulty, they opt out. Yes, there may be some increase in opt-outs, but we do not know.
If there is a tendency to opt out—also, what we are suggesting will only be a small extra cost to the employee—there may be some alternative approach, so that rather than opting out of the whole of automatic enrolment, they can opt down, perhaps maintaining the employer contribution but have a bit less from the employee. There is that.
Finally, on your third point, is it too late for people in their 50s and plus? It is pretty much too late, actually. Many of them, as Chris was saying, are people who had access to DB pensions and have also benefited from owning properties, seeing those properties increase in value, so they have more resources at their disposal—but not everyone.
Something that people often forget is that, when we look back to the great times of defined benefit pensions, less than half of the working population was ever in one at any one time, so a lot of people were not pension saving. What can we do for the over-50s? The state pension plays an important role—so it is important to ensure that that achieves a minimum—and there is the other option of carrying on working and working for longer. Fifty per cent. of people say that they like going to work and that they choose to go to work. So, there could be things around the labour market to make it easier for people who are older to work part-time or maybe work from home—I gather that is very popular at the moment. Those are things that can help smooth things for the older generation.
Q12 Sir Desmond Swayne: Anyone got anything to add?
Rob Yuille: Yes, quickly, one point about supporting people in their 50s who are in work and especially those who fell out of the labour market during the pandemic—it might have been to draw on a pension. They will be hit by the money purchase annual allowance. If they put more than £4,000 a year in, they will lose some of it in a tax charge. We think that that is hitting the wrong people. It does not need to be at the threshold it is; it could be higher or scrapped altogether, and replaced with tax recycling rules, because that is the behaviour it is trying to avoid.
Chris Curry: If I could come in quickly on whether increased contributions will lead to higher opt-outs, this is where it is really important that we consider the balance between employer and employee contributions.
As Nigel said, we have reversed how much comes from the employer and how much comes from the employee. Ultimately, if you look at it in purely economic terms, something the Pensions Commission did back in 2005, it makes no difference to the employer where the money comes from, because generally if there are higher employer contributions it means they have lower wage payments, so there is a trade-off involved in doing that, but it makes a very big difference to whether people will opt out or not.
For example, if you kept the current contribution for employees at 4% and had all the increase, up to 12%, from employers, so you had an 8% contribution from employers, if you opt out, you will get very little money back, but you would lose a lot of money from your employer. If you have it the other way round, so you keep employer contributions low and employee contributions go up, then it is much more likely that people will opt out because the amount of money they get back is much bigger and the amount they are losing from their employer is much smaller. So, the way you deliver these changes in policy is really important.
Q13 Debbie Abrahams: Good morning, everyone. I want to briefly revisit pensioner poverty. What do we know about the geographical distribution of pension poverty?
Chris Curry: We know that pensioner poverty varies by location. Some of the work we did in the adequacy report published last year—we can send more detailed information on this to the Committee—showed real issues around people trying to achieve living standards particularly in London, where living costs and housing costs are high and where a lot of income is low paid. Perhaps not surprisingly, pension poverty tends to be lower in the south-east, where incomes are higher on average.
Now, that is talking about poverty and that is not necessarily the same as reaching things like replacement rates. Sometimes it is harder for those with higher incomes to reach replacement rates because the target they are aiming for is that much higher than in general pensioner poverty terms. London is generally one of the hardest-hit regions.
Q14 Debbie Abrahams: Does anybody want to add anything to that?
Carl Emmerson: On pensioner poverty more generally, in some ways both state and private pensions provision have been a huge success over many years. Now we have a situation where, on average, pensioners are not more likely to be in poverty than the working-age population, whereas if you went back to the first half of the 1980s that really wasn’t the case. We had very, very high rates of pensioner poverty, so it is important to remember that pensioner poverty has fallen a lot and pensioners are not, on average, a group that is more likely to be in poverty anymore.
Having said that, it is important that we do not get complacent. Going back to my first answer about the Pensions Commission and if we need another review, I worry that, in some ways, the recent success of pensioner poverty being pretty low and average incomes of recent retirees being pretty high, might make policy makers a bit complacent and think, “Oh, it’s all fixed. We don’t need to do anything.”
Pensions policy is actually a long-burn game, and we are already discussing what we can do to help people in their 50s and if it is too late. Maybe, we should be worrying about the poverty prospects for people in their 30s when they get to retirement and trying to fix that now.
Q15 Debbie Abrahams: But, as you rightly say, we are talking around there. Overall, we are seeing an uptick of pensioner poverty, but I take your point that compared to previous decades, and compared to working-age and child poverty, it is much lower. We are still seeing women, particularly single women—one in five—living in poverty, aren’t we?
We talked about the labour market being a key issue, not just in terms of access to it but also progression within it. Has anyone done any scenario modelling about what we could expect over the next decade, without there being any interventions, in terms of pensioner poverty?
Chris Curry: Trying to project and model pensioner poverty is incredibly difficult, mainly because the way in which it is measured in the UK compares the income of people in retirement to the income of the working-age population. It requires us not just to project the incomes of people in retirement or the people coming up to retirement—the 50 to 65-year-olds—but also to model the rest of the population and the economy, to work out what we think average incomes are likely to be.
We have done work on this and I know that the IFS has done work on this in the past, and I think it has been broadly in the same ballpark. We projected the decrease that has happened recently. It is very hard, especially in the pandemic situation, to know what is likely to happen with incomes generally across the population in order to be able to generate poverty projections. However, we have done work around what we think is likely to happen to female pensioner incomes and these other under-pensioned groups. We are seeing exactly the same issues that historically have always been there remaining with women.
The state pension underpin is improving and getting better, and we expect to see women getting better state pension provision increasingly year on year, until I think it is not too long—from memory, it is somewhere in the late 2020s—before a woman reaching state pension age will have the same expectation of a state pension as a man reaching state pension age. That is quite a big shift from how it was not that long ago.
We are also seeing that the amount of pension that women have accumulated up until now and their potential to accumulate further between now and the time they get to state pension age, or the time they start to take their pension, is still reduced compared to men. The majority of that does come through the labour market, through lower pay and lower hours. In some ways it is exacerbated by the pension system, in the way in which, for example, the parameters around automatic enrolment deliberately do not include low earners, but also perhaps inadvertently exclude people whose earnings are high enough for them to be automatically enrolled but maybe across more than one role. So, they are not being automatically enrolled even if their earnings are high enough that if you took them together, they could be.
There are things that can be done around the margins but, as I said earlier, I think a lot of the differences that we are seeing in pension outcomes, especially for women and for other disadvantaged groups, are driven by labour market participation.
Q16 Debbie Abrahams: I have a final question, very briefly. On international comparisons—I think there was an EU report last year at the same time, looking at adequacy—are there any lessons we can learn from them?
Chris Curry: Again, it is difficult to say, but I think that international systems tend to do well for women, in particular, where they are more state-based. I don’t think there are many international systems where there is large-scale private voluntary saving or workplace-based saving where women fare as well as men.
Q17 Debbie Abrahams: Any other contributions from the panel?
Carl Emmerson: Relative to other European countries, I would say that, on the one hand, we spend a lot less on our state pensions but, on the other hand, our state pensions are very flat-rate. Not spending very much is clearly one way of saving costs, but it will be one way that will push up poverty. Making it flat-rate will target resources and actually reduce poverty relative to systems where you give more to people who have paid in more and who are richer.
If your objective is poverty reduction, the UK has one weakness, while we have a relatively low-cost system. It has one big strength in that we have moved to this flat-rate system that is, in some senses, much more generous to people who earn little through their lives relative to people who earn a lot through their lives, because they pay in very different amounts but they get the same out.
Nigel Peaple: I might just add one point on female pensioner poverty, thinking about couples and recognising that there may have been different involvement in the workforce. Let us say that the man has worked for more years than the woman, perhaps because of child caring or looking after aged parents. A key issue is that, as a household, they have developed a certain level of income, but there is still a divorce in, I think, one in three cases, so is that pension being shared evenly between them? I am sure you have seen other policy initiatives’ discussion on this issue, about making pension sharing on divorce more of an automatic element. That could be another way of dealing with female pensioner poverty that isn’t about the money in, but more about where the resources come from.
Rob Yuille: I was going to say exactly the same thing. One more point is that one of the structural reasons is the gender pay gap, and that is different from a career break. There should be more support for carers, and that comes back to your question about people in their 50s leaving the labour market.
Q18 Chris Stephens: Nigel, if I could start with you, I want to look at guidance and advice. More than half of the accesses of defined contribution pension pots are being carried out without impartial guidance or regulated financial advice being used. What more should the Government and public bodies do to ensure that people get the advice and guidance they need while saving for retirement?
Nigel Peaple: This is about decumulation of the pots. I think that, on the one hand, some of the proposals from this Committee at the last stage were very welcome. The idea about limited advice and enhanced guidance was helpful. There was also the suggestion that Pension Wise should at least trial the idea of automatic appointments, and that sounds like a good idea, too. But, as you may be aware, the PSLA’s membership of workplace pension schemes have been thinking about how to get good outcomes in retirement—not so much about the advice and guidance, but about how to get good outcomes. That is why we have suggested that, really, we should have a thing called guided retirement income choices—the idea that there would be a product or a solution at decumulation that would meet all of a person’s needs, through a mixture of some cash, some invested fund that can be drawn as appropriate and then a guaranteed income for later life.
In that world where we envisage that you would have product standards, or standards for these solutions, we believe that you would not need to have the same level of extremely sophisticated and expensive advice as is generally felt to be necessary at the moment, because you would have had an employer and the trustees of pensions schemes choosing a product that is basically fit for purpose and is going to be kind of okay for the person. It is a bit like when you go to the car dealership and buy a car. You do not have to understand everything about an engine to get one that works; you rely on the fact that it has passed various standards, including the MOT.
Q19 Chris Stephens: Thanks, that is very helpful. Rob, pensions dashboards appear to be at least two and a half years away. What should bodies such as the Department for Work and Pensions, FCA, MAPS and the Pensions Regulator be doing before dashboards arrive to help to improve engagement and understanding among non-advised savers?
Rob Yuille: I think we can build on the recommendations that the Committee made in its previous report, which I also endorse. Some of the same issues that you picked out in terms of accessing pensions also apply when saving for retirement, but with some important differences. Pension Wise is a great service with very high levels of satisfaction. It is specifically about accessing pensions, so in addition, we should be looking at encouraging people to use MoneyHelper services more broadly, as well as enabling providers—pensions schemes, employers and others—to do more, building on the recommendation you made about enhanced guidance, by encouraging people to save more, for example.
Depending on how that is done, if it is based on someone’s own financial situation, it might look like a personal recommendation, which is regulated financial advice. We would like the Treasury and the FCA to look more closely at that, and again, they can draw on some recommendations that you have made. Similarly, depending on the format that they use—email versus letter—it might be construed as marketing. Again, that is something that could be looked at.
We should not rule out regulated financial advice in the savings phase either. It is probably more important when making those complex decisions about accessing pensions. As people get closer to retirement and have many different pensions to draw on, they can benefit from regulated financial advice. We should look at all those options.
Q20 Chris Stephens: Thanks very much. Chris, you will be aware that the Committee recommended that the DWP explore the pensions dashboard programme and whether dashboards could get aggregated information about individuals’ savings to help regulators ensure that people get the guidance they need. Could that work? Are pensions dashboards alone enough to help people understand how to save for and draw down their pensions?
Chris Curry: I am going to momentarily switch hats, if that is all right, and speak as principal of the pensions dashboard programme rather than as director of the Pensions Policy Institute.
On the first question, I am afraid that the recommendation from the Committee, or the idea that information from pensions dashboards can be aggregated and shared with the Pensions Regulator, the FCA or other Government bodies to help policy development, is not something that will be possible with the pensions dashboards as they are currently being set up. The main reason for that is the emphasis on security and ownership of the data in pensions dashboards. We think it is critical that people are completely confident that their data, which is currently stored with their pension providers, is being seen only by themselves and nobody else—including regulators or Government Departments, however well-meaning. We need our dashboard system to operate in as safe an environment as possible.
We need consumers to be incredibly secure and confident that their data is not going anywhere it should not be, so the system is predicated on it being where it currently resides with a provider and momentarily displayed on dashboards until that individual decides they want to use the information somewhere else. It will not go anywhere else outside the system. It is not retained or stored anywhere, so there is no way that it can be accessed, which means it is not possible to use it for those kinds of policy or research purposes, and also it is not available to be hacked, scammed or used by anybody else. The design of the system rules out that kind of data sharing, so it is not something that would be possible.
In terms of dashboards alone being enough, we have always been clear that dashboards are a really important step, and probably the first step, in making sure that individuals can understand what they currently have, find pensions they may not realise they have entitlements to and ones they did know about, get a reasonably current up-to-date value so they can see what they are worth, and be able to bring them all together and see them. We hope that in itself will generate interest and some desire to understand further.
There will be signposting available in pensions dashboards to help people move on to something else—the MoneyHelper guidance service or independent financial advice, to help them make further steps. We have always been clear that, by themselves, all pension dashboards will do is to help improve people’s understanding of their own position. We know that is not necessarily enough for them to take action or to do things differently. We know we will have to work with the whole ecosystem on pensions dashboards, potentially a whole new set of arrangements and maybe new policy from the DWP, the Treasury, the FCA or regulators, to help increase people’s financial confidence and capability and make it easier for them to take the information they get from dashboards and turn it into something concrete for them.
Q21 Chris Stephens: Thanks very much, Chris. Back to you, Rob. The ABI says it is unclear whether signposting to guidance or advice services on dashboards would be considered an advertisement and be banned. How and when can we resolve that issue so that dashboards can play a role in increasing take-up? Could you tell us whether any formal legal advice has been received on this matter?
Rob Yuille: I am not sure it is necessary to take legal advice. It should be resolved quite easily. It is similar to the concern I mentioned that enabling someone to take action is precisely what dashboards are there for. Chris and colleagues in the regulators will need to find the balance between making sure it is not used as sales and encouraging people to take action, whether that is saving more, transferring a pension or some other action. There are live consultations at the moment. The FCA will later consult on how it intends to regulate dashboard providers. Those are the ways in which it will be resolved. I am confident that it will, but it is an important point.
Q22 Chris Stephens: Thanks. That is something we will certainly look at. Back to you, Chris. You will be aware of the recommendation that the Government trial automatic Pension Wise appointments at 50. Is there any good reason you are aware of why the Government have not progressed on that?
Chris Curry: That is a matter for the Department for Work and Pensions and the Treasury together. Referring back to something Rob said earlier, the current Pension Wise appointments are very specifically targeted and limited in what they can cover to a specific area. As was also pointed out, the people who go through them find them incredibly useful. The customer satisfaction scores are generally very high.
I think there is a question of whether it will be something that is not quite compulsory but that everybody is automatically enrolled in, and whether the current type of session is exactly the right one. If someone is not necessarily thinking about accessing their pension, do we want to enrol them in a session that tells them how to access their pension? Is that something that we might want to look at in conjunction with other potential policy interventions, such as the mid-life MOT or something like that, and talk about it in a more holistic way, potentially? There are a number of potential advantages, and I think the power of automatic enrolment is really important. Without knowing, I am assuming that the Department and the Treasury are looking at what the potential outcome would be.
It was interesting that in the stronger nudge work, although there was a positive impact, it was a relatively small one. It is a balance between what we are trying to achieve and whether what we have already got is the right tool for meeting those objectives.
I will briefly come back to Rob’s previous answer about dashboards and whether advertising and other things will be allowed. As well as the FCA consultation, the Money and Pensions Service will be consulting over the summer on things such as display standards for pensions dashboards. There will be an opportunity for the industry to participate in that discussion, so that we can make sure we get the balance right between allowing people to do something positive with their pension and avoiding them inadvertently doing something to their detriment.
Q23 Chris Stephens: That is very helpful. Nigel, these are my last questions to you: are schemes and employers doing what they can to help people access guidance and advice? If not, why not? Also, how much of a solution will this Committee’s recommendations for new definitions of enhanced guidance and limited advice be for people?
Nigel Peaple: Many employers are doing a great deal within the workplace to engage their staff with financial wellbeing generally, including pension saving, although they sometimes struggle with that. There is a lot of concern about falling within the regulatory perimeter: comments from the HR department about what you should do on pension saving might accidentally become regulated advice. That is an issue that has been bubbling around for a while. The PLSA produced a guidance document on that last September to give employers an idea of some things we know they can do, with some things they probably don’t want to do. So, they are doing things, but they are operating within a context where they are worried about slipping into regulated advice. They are worried about accidentally taking part in any financial promotion, which is a regulated activity.
They are doing stuff, but they could do more if things such as the Select Committee’s proposals around limited advice and, in particular, enhanced guidance were adopted. It might also be worth looking into the definition of financial promotion, to make it easier for an employer, or the pension provider operating on behalf of the employer, to be allowed to talk about pensions in the workplace within some set of reasonable parameters, such as, “What’s your target income? Are you saving enough?”, rather than issues that are more controversial, to my mind, such as transferring out of DB and all the risks that that involves. There is more to be done there, but I think there is some potential.
Q24 Dr Spencer: My questions are mainly on retirement living standards, and they are to you, Nigel, as you would expect. The living standards you have set up are now going out to 15 million people. What is the impact of that? How is it working out?
Nigel Peaple: The first thing to say is that they have been up and running for only two years, so it is a little early to have a really substantial set of evidence. However, we are finding that we have, frankly, been agreeably surprised by how much they have captured people’s imagination, and how much the wider pensions industry and savers have adopted and starting using them.
We have been asking pension providers and schemes about what sort of uptake they have got, and I have a few little stats from this relatively early period. For instance, when one pensions provider sends out information on pensions, it normally gets about a 10% open rate—just 10%. That is why I always say automatic enrolment works better than engagement. However, it found that when it started using retirement living standards, that rate went up to 30% or 40%, which seems like a really encouraging uptick in interest.
These are very specific, but another similar example is that from changing the way it talks about retirement living standards—often using fintech tools on the company website, as well, by the way—they have found that people started saving more. I haven’t got a proper study yet showing how effective the standards are, but it is clear that they capture people’s imagination.
Just two more things: when we’ve made changes to the standards, we have found that there is an enormous amount of interest in the media and from readers and listeners who are full of questions, interest and engagement. Just to come back one level: when we originally came up with retirement living standards as an idea about four or five years ago, it came from our member pensions scheme saying that the trouble is that nobody really has an idea of what to expect in retirement and that nobody knows how to engage with it. All you see in the Sunday papers is lots of people dressed in white and on yachts in the Mediterranean. That is not what most people’s retirement is going to be like. So we wanted to find some way to bring it into reality and make it engaging, and also to be able to educate the conversation across the country. That is what the retirement living standards are about.
The very fact that people are a bit more engaged with retirement living standards and are thinking about their pension, and are not just assuming, because they are in a workplace pension at 8% of banded earnings, they are going to be okay, that is a good thing—realising that the money you put in will affect the level of pension you get out.
Q25 Dr Spencer: Thank you for your comprehensive pitch on the way you put it together. One of the criticisms about the standards has been that it is not necessarily clear what you have to do to achieve those living standards, in terms of how much you have to put in. What would you say to that?
Nigel Peaple: Obviously, we considered whether we should promote particular savings levels when we came out with the retirement living standards. You could say, “Therefore, to achieve this level, you need to save at x throughout the whole of your lifetime.” At the time, and it is still the case, we decided not to do that bit, because so much depends on how much savings you have got already, what age you are, what your other financial circumstances are. That is why we said, “We share with you this research from what the UK population says about three levels of income, and you can engage with it,” but then it is for others—pension schemes, advisers, others—to be able to convert that into some simple rules of thumb.
The way that fintech is developing now, quite a lot of the large providers, if you go onto their websites, have stuff to engage people about their saving. They say, you know, “Where are you now? What do you fancy in terms of your lifestyle? Do you want to do a few more holidays? A few less holidays? More trips to supermarket? Fewer trips? Which supermarket?”—that is a very important point—and, “This is what you would need to do to achieve that.” You might need to put some more money in. Occasionally, they might say, “You’re on target”, but that doesn’t happen very often.
Q26 Dr Spencer: Is there a risk that people might look at it and feel that achieving that is unattainable?
Nigel Peaple: This is very interesting. I think it is partly about framing. For instance, when I talk about this, I always say, “The state pension will get most people to that minimum level.” If you are on median earnings and you are saving over a lifetime, you will be well on the way between the minimum and the moderate level. If you are a couple, because you have shared earnings, you are well up to moderate level. Partly, it is about framing. I think there may be a case for changing the labels, so that it could be more clearly seen that the minimum is not just the minimum—actually, it is okay; you can function in society—and that the moderate is really quite pleasant, and it is what most people live on in the UK, and that the top level is actually the top; it is probably for 20%.
Q27 Dr Spencer: Should it go on the dashboard?
Nigel Peaple: We have been suggesting and requesting that it be on the pensions dashboard for years and, at the very least, that there is some clear set of targets on the dashboard or accessible from the dashboard to help people with their retirement planning because, as Chris just said, the dashboard alone—a series of numbers from your pension scheme, which you may not really understand—won’t really mean very much. We really would like it to be on the dashboard, and we hope the Minister will listen to this one day.
Q28 Dr Spencer: Chris?
Chris Curry: Sticking my pensions dashboard hat back on again, potentially at some point, they could be on the dashboard, but they won’t be on the first iteration. I think I am a little bit more hopeful than Nigel that people will understand the numbers well on the dashboard when they first go out. We are doing a lot of user testing and research to work out what people will understand and how they will see it.
Longer term—absolutely. The dashboard will continue to evolve. The dashboard will continue to be driven by what consumers want and how they best use it. There will always be underlying user testing and user research going on. I think the desire at the moment is to deliver something that is of use in a reasonable timeframe. Once we have that in place, we will then be able to test how well dashboards interact with what is already existing in the system—as well as the retirement living standards that the PLSA is developing, also some of the other tools and calculators that fintech are working on. We would expect the dashboard to be working very closely with a lot of those different functions and a kind of new ecosystem to be developing around the pensions dashboard. Whether that needs to be on the dashboard or is something that could be closely associated with it, that people then move on to when they decide that is what they want to do next, that is an open question, but we will be driven by what consumers understand, what they want and what we think they will find helpful, as to what exactly ends up on the dashboard at which point.
Q29 Dr Spencer: I have to say, Nigel, you have sold me. Before, when I have looked at my pension numbers, it is very uninterpretable, but just seeing what each breakdown is makes it more real in terms of thinking through.
Rob Yuille: If I could just add briefly, as Chris and Nigel said, it is information that is accessible from dashboards. One of the points that the DWP is consulting on is whether it should be possible to export data from dashboards so that you can use it in another application—so, “These are my targets. This is what I get if I retire at 60. What if I retire at 70?” You can only do that if you are able to use your own data with your own consent, so we hope that will be part of the dashboard solution.
Q30 Chair: Thank you very much. Can I ask you about pensions savings among the self‑employed, which we have mentioned once or twice? It is much lower now than it was 25 years ago. I would be interested to know why you think that is, and whether you think that, through ideas in the current NEST pilot or something else, this can be fixed without an incentive mirroring the employer contributions available to employed people. Who wants to comment on that? Carl?
Carl Emmerson: You are certainly right that coverage has fallen very sharply. If you go back to the late ‘90s, almost half of self‑employed people were saving in a pension. It was not much lower than the employed population. It fell almost continuously until about eight years ago, and it has now been running at just below 20%, so an astonishing change. We did a piece of research to try and look at what explains that, and to be honest, I expected it to be, “There are lots of people who are becoming self‑employed. They are very low‑income, they are only doing it for a bit, and that is where the drop will be.”
That is really not what we found. The drop in coverage was most prevalent among higher‑income self‑employed people—they are not that well off, but they are better off than other self‑employed people—and was much bigger among people who had been self‑employed for quite a while, seven years or more. It was long‑tenured and—in relative terms—better‑off self‑employed, because they were the ones who used to be in pensions and their coverage rates have fallen the most. It does not really answer the question, except to say that it is not all about people just moving into self‑employment for a short period of time now, and that used not to be the case, or lots of very low‑income self‑employed now, and that used not to be the case. That is not explaining it.
I think the NEST-style solutions can help; I think I would be looking at other things. For example, when people go from being an employee to being self‑employed, I wonder whether there is something smarter we can do at that point of transition, rather than the default being, “You stop saving in a pension.” Maybe we would like the default to be, “You continue to save at your employee rate of contributions, and maybe we should write to you and tell you what your employer would have been putting in had your income remained the same.” Maybe there is a moment there where we should try and capture people and stop the default being, “You stop saving in a pension.”
Maybe there are smart things we can do around tax returns by telling people, “You’re self‑employed. Did you know that an employee on the same income would have put this much into their pension? Would you like to do that now?” That might be a good moment, because it is also a moment where you can say, “By the way, you will be getting tax relief on that contribution instantaneously, because you are now filling in that tax return.” There are certainly other moments we can look to that would be based on what we have learned from automatic enrolment, and using that and trying to harness it to boost coverage among the self‑employed.
Q31 Chair: Anyone else want to comment on this? Chris?
Chris Curry: I kind of agree with a lot of what Carl has said, and we did think about this when Carl and I were part of the 2017 automatic enrolment review. It was quite an important area of discussion, and there were planned to be pilots run by the DWP, but there was nothing that really came back with anything that was too positive in terms of ways to get the self‑employed saving. Everything we have looked at has some drawbacks. With the last idea that Carl was talking about, using the tax system, there are some really powerful incentives; but equally, when the self‑employed are using the tax system, they end up having to pay a tax bill at the end of it, so the idea of them also making a lump sum contribution to their pension at the same time makes it more difficult to see how it would work in practice. There is never really an easy solution.
To pick up on a couple of the other things that Carl said, the self‑employed are not a single group. There are so many different types of individual who are bound up within the self‑employed, and there is not just a single problem, as Carl was saying. With the ones you might traditionally assume are fine as self‑employed, their pension provision has been falling as well. There are people who are low‑income self‑employed, and there are people who are self‑employed for parts of their career, not all their career, so we need to have a look at how that interaction works.
One thing the PPI found in the research that we did as part of the automatic enrolment review was that savings overall for the self-employed group were not necessarily that different from people in employment; they just saved in different ways. Their pension saving was much lower but, for example, their housing wealth was higher. That was not universally distributed, obviously. There is a real distribution there, and some people have lots of housing wealth and some people have none.
Part of the thinking might have to be alongside trying to encourage the self-employed to get into pension saving. The same with an employee—the earlier you do it, the better it is going to be for people. The more of a habit it becomes, the less you have to worry about it and the less you miss it. There must be something behavioural in there.
It is how it fits in with their other forms of saving, which goes back to how we were talking at the start about how to measure adequacy and what is important for people in retirement. Is it just pensions for the self-employed? Is it something other than pensions, or something to supplement pensions, that we need to think about, in order to look at it in a more holistic way?
Q32 Chair: Talking about different groups within that category, I want to ask you about people in the gig economy. In the Uber case last year, the Supreme Court designated Uber drivers as workers, and they are now being auto-enrolled into a pension. Some of us had an email—I think this week—from Hermes saying that they are now starting to pay pension contributions for their drivers. Do you think the law needs changing to clarify the position for people in the gig economy more widely?
There’s another specific point. The Prospect trade union says that the optional three-month delay before auto-enrolment disadvantages freelancers. Is that an example of something that you think needs to be changed to deal with the growth of the gig economy?
Rob Yuille: I can start, if you like. As well as the Uber case, the Pensions Regulator’s response to it was helpful and encouraging, and made it clear that firms like that need to fulfil their automatic enrolment obligations. Whether legislation is needed is a broader question than pensions. The risk is that creating a new category of workers would cause other forms of quasi-employment to spring up.
What the Pensions Regulator came out with was impressive. I was also struck by Uber’s suggestion that there should be a cross-industry scheme, because there is the same multiple jobholder problem. A driver might have the same cab but several different apps that they are working off. Ideally, they would work together, but that is easier said than done. It would be interesting to explore that. Some of the same points that we have been mentioning all along are wider more structural issues that need to be addressed.
Nigel Peaple: We haven’t looked at it since the Uber decision, but prior to that we were of the view that many gig economy workers would need to be redefined as workers, and that did seem an elegant solution. I’m happy to go away and look further into this issue to try to work out what could be done.
Rob Yuille: The point about the waiting period could be looked at, but the earlier points about the need for a long-term strategy and sequencing in the right way would again run the risk of freelancers having minute pots from different employers. Once qualifying earnings have been reduced, that might be a more workable prospect.
Chris Curry: It isn’t research that we have done, but it might be worth looking into how many employers use the waiting period. Speaking as PPI, as an employer, we don’t use the three-month waiting period; we automatically enrol as soon as people start. I guess that depends on employer behaviour, as well as what the law says about the potential waiting period.
Q33 Siobhan Baillie: Thinking about changes in scheme design that would most improve outcomes, Carl, the IFS has said that although schemes introduced since auto-enrolment operate at low costs, that is not the case for legacy schemes. What do you think the Government should do for legacy scheme holders, who tend to be older?
Carl Emmerson: We took some data provided to us by a fintech company, which had information on people’s historic pensions. In those data we could see what they were charging, what returns they achieved, and also how they were invested. Those were predominantly people in their 50s and quite a wealthy group on average. Pensions they had taken out more recently, on average, had lower charges than ones taken out longer ago. We found no evidence that the returns that they were achieving varied across whether the pension had been taken out recently or longer ago, so it wasn’t the case that the more expensive schemes were delivering better performance—at least, in recent years.
We also looked at how much was invested in riskier assets and how that correlated with age. If you look at pensions taken out recently, there is quite a strong correlation with older people having safer pension pots, as you might expect. If you look at pensions taken out a long while ago, there is no correlation there, so it doesn’t look like they’re as well invested for individuals’ current situations. The same is true when people were asked about their attitudes to risk; that correlated with how the pensions were invested if they were taken out recently, but not if they were taken out longer ago.
It does look like older pensions are much more at risk of being too expensive and badly invested, given, perhaps, someone’s current situation. It suggests that we need to do more—regulators need to do more. We need to do more to engage consumers to get people to have a look at their pensions and think about changing how they’re invested—perhaps move them. The current environment, where new pensions taken out are typically low-cost, hasn’t solved all of those legacy problems, even though many of them have been reduced.
Q34 Siobhan Baillie: Awareness is obviously an issue, in terms of the older pension. When people were aware of the kinds of differences, were they acting? Did you see a change?
Carl Emmerson: That wasn’t something that we could look at with this data, because it was a set of people who’d contacted this company and said, “Please can you track my old pensions?” The company was then providing them with this information, and then looking to engage with them and charge them in order to provide a service.
The situation, potentially, could be very different for the rest of the population who are not engaging with those kinds of outlets. One imagines that it could go either way. It could be the people who are kind of aware that they have got a big pension from the past that they need to do something about who are reaching out. It could be that that group is much more engaged, and that you’ve got a bigger problem among the population who really aren’t engaging.
Q35 Siobhan Baillie: Does anyone else want to contribute on legacy pensions?
Rob Yuille: Yes, I am happy to. It has been an area of focus for regulators for quite a while. The FCA did a review of treatment of long-standing customers around five years ago, and the guidance is very well embedded now. Providers will do regular product reviews to ensure that they provide value for money. The FCA is consulting on a consumer duty, as well as having just consulted on value for money, jointly, with the Pensions Regulator, which is encouraging. There are several measures in place on value for money.
In the example that Carl gave on the research, by definition, those people have acted because they were shopping around and looking for a new home for their pension. That is critical. I will, very tentatively, disagree with the IFS: it is not necessarily the case that if someone starts to de-risk their pension as they approach retirement, that is good, and if they remain invested in riskier assets, that is bad, especially if they are intending to stay invested throughout retirement. The critical thing is that people pay attention to their pensions. There is a very busy non-workplace pensions market, with lots of new entrants and a lot of activity.
Q36 Siobhan Baillie: With pensions, it always comes back to awareness and knowledge, doesn’t it?
Nigel Peaple: I have one small point to add to this. My understanding is that pension providers, and the insurance industry more widely, have done quite a lot of reaching out to holders of those legacy pensions with the high levels of charges. However, part of the problem is cases in which they do not have data on people because they have moved around and not updated their pension provider. I think there is a sub-group of people who have lost contact with their pension.
It may be—goodness knows what the journey will be like—that, as the dashboards get up and running over the next year or two, we will see further ways of enhancing data and reconnecting with people, and that might help to solve those other problems.
Rob Yuille: I have a couple of points to make on that, if we have time. To go back, partly, to the piece of guidance that the FCA put out, our members have done a lot of work to invest in tracing services. Tracing is quite good, but the challenge that they find is that when they think they’ve found the new address of the customer, they’ll write to them to confirm it, and the reconnection rate is not as good.
We would like industry to be able to say to Government Departments, “We think this is the right person. Can you confirm that—yes or no—using Government data in a secure way?” There is a wider Government data strategy, and we think this could be explored as part of that.
Carl Emmerson: In some ways, the version of automatic enrolment that we have introduced in the UK will make these issues worse and much more important, in that you work for an employer and are enrolled into a pension, and then you move to another employer and are enrolled into another pension. You get lots of small pots, and it is much more important that you keep a track of all those. It is putting quite a big burden on the pensions industry, which is having to administer lots of those small pots.
I was struck by the fact that Ireland looked at that and decided to go for the opposite approach, whereby individuals get one pot. That puts more burden on employers. At the IFS, we have about 70 people on our payroll, and we enrol them all into one pension. In the Irish model, you would have to enrol them into their pension, which could be with 70 different providers, but it would be much less burdensome on the pensions industry and perhaps easier for individuals to keep track of where they are at. In some ways, we have baked this in with the version of automatic enrolment that we have gone for. We either have to live with that and try to tackle problems of engagement and connection with old pensions, which is made much more important, or we do something very radical and change the system. That would move the burden much more away from the industry and individuals, and much more towards employers.
Nigel Peaple: That is a major issue. It is important to recognise that the model in the UK, which is employer and workplace-based, has the advantage of keeping the employer bound in and committed to the pension. Historically, that is how we have done things in the UK—with a contribution from the employer. That makes achieving adequacy targets much easier. We don’t think it is a good idea to move away from a model in which the employer plays a key role. Instead, it becomes more like bank accounts. You can imagine having some minimum statutory requirement, which probably will not be enough.
Rob Yuille: The small-pot challenge that we face is an example of where industry and Government have worked together. That is ongoing. We are doing a project with the PLSA involving many others from the sector, and I am happy to provide more information to the Committee as that progresses, because it is quite relevant to this inquiry.
Q37 Siobhan Baillie: Rob, I want to ask you about the ABI. I know you have said that you would like to see the Government and industry foster a culture of value for money in which investment returns are as important as price. What we are interested to know is: how that does not undermine the charge cap, which has been so effective in driving costs down? What are your views on that, and do others agree?
Rob Yuille: It should not undermine the charge cap. Of course, price is an important part of value for money, but it is not the only part, and we are pleased that the Government and regulators agree. The FCA have recently consulted jointly with TPR, in line with the Committee’s recommendation, on metrics for value for money. They categorise the three main measures as investment performance, cost and charges, and customer service. We think that is broadly right, although there is a lot in that third customer service bucket, including things that we take for granted but are really important, such as quality of governance, quality of admin and things that people value, such as whether there is an app or whether transfers are done quickly. We are encouraged by the direction of that.
Siobhan Baillie: Anybody else on undermining the charge cap?
Chris Curry: We have done some work with the Pensions Regulator in this space, and I think it is along very similar lines to what Rob was talking about. Value for money is probably the most important aspect. When you look at the whole performance of a pension outside contributions, it is probably the thing that determines outcomes the most in pension provision. All the work that we did suggested that, internationally, you might want to separate administration and governance from customer service, so that you have a much more focused engagement level—thinking about the customer, but making sure that things are administered and governed properly, because there would be problems if that did not happen. The examples that we have seen in Australia demonstrate that.
There is always a trade-off between wanting to get the best overall net return and net performance. For that, you need to look at the investment return and investment performance, which could be managing volatility, looking for the right return, and getting the right set of outcomes, and the right price and charges for doing that. Research that we have done over the past two to three years suggests that it can be feasible to do that without undermining charge caps. I think there are examples in different parts of the industry where that is happening, so I do not think it is an either/or. I do not think you have to say, “Well, we can expand investments and look at different ways of getting better returns for our members, but we can’t do it within the charge cap.” You can try to balance that and make them both work.
Nigel Peaple: The average charge rate is about 0.4% or 0.5%, so it is well below the charge cap point of 0.75%. My other observation is that it is true, particularly in trust-based pension provision, that the trustees are very focused on keeping costs very low. It is probably good that they be encouraged to think a bit more about value for money, rather than just the cost going in, because hopefully you will get better things coming out of it, such as better service and engagement for the savers, better investment returns and so on.
Q38 Chair: Can I raise a couple of points with you, Nigel? We talked earlier about gender pensions issues. PLSA has made a couple of suggestions in this area. Can you tell us about them? One is that you said it would now be possible to auto-enrol people with several low-paid jobs, who do not earn enough from any one job to be auto-enrolled. That picks up on a point made by Chris. I am interested in how that could work. The other point is that you have supported encouraging employers to maintain pension contributions during maternity leave. How practical and feasible do you think that is?
Nigel Peaple: On the first of those points, the issue about people earning more than £10,000 a year but falling under that threshold because they do part-time jobs is an important one, and it affects women more than men because they tend to do more part-time work. It seems crazy that there is this group of people who are missing out. We think that the solution is probably working through the tax system in some way—talking with HMRC about Making Tax Digital.
It may be, and this is a bit of a Rubicon, that as a result of the dashboard, in a few years’ time you have a more collective view of a person’s overall income and savings, and you might be able to find a way to encourage them. We do not have a specific proposal in this area, but here we are. I like to tell my team, “We are in the 21st century,” and they laugh at me and point out, “That started 22 years ago.” It does feel that given all the developments in digital, tech and tax, it ought to be possible to re-grapple with this and find a solution on the multiple pots.
On the question of employers paying pension contributions during maternity leave, why not, really? This is a big societal issue, and it is about how society treats its ability to reproduce and how it helps different peoples, genders and generations. Philosophically, it seems entirely justified and reasonable in our social and political view, but the key will be getting employers on board. I know from other conversations that I take part in on things like the living pension, which is linked to the living wage idea, that some employers are very committed to trying to improve pension provision, and I think it is really worth pursuing.
Q39 Chair: Anyone else have any comments on either of those points?
Rob Yuille: On the gender pension gap more generally, there are some similar themes to a lot of what we have been saying. There are some things that we can change in the pension system to help, such as making pension saving work better for lower earners, and sharing wealth within households better. The extreme case of divorce is one example, but also a non-earning spouse can contribute to a pension and get tax relief on it. The threshold for that is quite low and has not increased for a long time. I do not think we understand enough about how income and wealth are shared within households, either. There are things that can be done within pensions, but a lot of the challenges are broader, structural factors to do with the labour market. That does not mean that groups such as your Committee should not explore them.
Q40 Debbie Abrahams: Can I ask one final question, Chair? Do you think the state pension age recognises that we are seeing a decline in life expectancy for women? That is a national issue. We have known that since 2015. Certainly, life expectancy is also decreasing in deprived areas. Is that recognised? I think it was Nigel who said quite early on that state pension age reform should continue, and I was not sure whether you meant that the increase in state pension age should also continue.
Nigel Peaple: Shall I chip in with a few thoughts on that? I do not think I said that the state pension age should change—that might have been one of the think-tankers.
Debbie Abrahams: No, you said state pension reform.
Nigel Peaple: Women’s longevity is still higher than men’s, so arguably the current compromise—whereby both men and women have the same state pension age, rather than our adjusting it—is reasonable.
Separate from that, you raised the fact that there is very different longevity in different parts of society and across the UK, so some people say that you should lower the state pension age for certain parts. There is a consultation going on exploring whether there should be a postcode linkage to the state pension, which I think, in practical terms, could not possibly work, because everybody could move to the area with the lowest state pension age.
I do not think it is feasible to alter the state pension age downwards in any way. Hopefully, it will not go up too much more, because of the issues that you have talked about.
Carl Emmerson: For women, the state pension age has risen from 60 to 66 in recent years. That is a big increase. It is still the case that a 50-year-old woman today would not expect to receive her state pension for less long than a 50-year-old woman from 50 years ago, say, because of the increases in longevity at older ages over that period. That gives quite a strong case for having increased the state pension age.
That comes back to what I said earlier: we should probably think more about how the system works for people who are just below the state pension age. If the state pension age is increased, it exposes you to the working-age benefit system for longer, and that is much less generous than the state pension system. There has not been enough thought on how universal credit works for people who are in the last few years before they reach state pension age—what work search requirements should we put on people, and how do we do the health test? Should those things be the same as they are for the whole of the working-age population, or should they be different?
At the moment we have this big bang change when you reach the state pension age: you get a lot more money, and you do not have to seek work. Do we want a more gradual thing occurring before then? That would help many of those who, when the state pension age is increased, perhaps want to work but are not able to—perhaps they cannot find a job or their health means that they cannot work. That is the group I worry most about. The solution is probably in the working-age benefit system, not in the state pension system.
Debbie Abrahams: That is very helpful indeed, although I must say that in my constituency women are now living two years less than five years ago.
Chair: Thank you all. This has been a very helpful and useful session to start our inquiry. That concludes the questions we wanted to put to you. We will carry on our meeting, because we are going to discuss, in the light of what you have told us, who else we would like to hear from in the course of the inquiry. Thank you very much for joining us and giving very helpful answers.