The Chairman (Mr A. G. Lowe, F.F.A.): Welcome to the Institute and Faculty of Actuaries (IFoA) sessional event on why the pensions dashboard is more than a data collection exercise.
My name is Andrew Lowe and I am the chair of the Future Pensions Landscape working party. I qualified as an actuary in 2005. I have done consultancy work in the public and the private sector. I am now the director of change and data solutions for Equiniti.
With me today is Michael Roe, who is the development director at Origo. Michael has been working with various parties across the industry who are interested in getting pension dashboards off the ground.
Now I shall hand you over to Michael to set the scene about the pensions dashboard.
Mr M. Roe: Origo is a local FinTech company. We are based at Edinburgh Park on the floodplain. There are about 70 of us. One of our claims is that we are one of the oldest FinTechs in Scotland.
We have been involved with the pension dashboard from the outset. We are technologists; we are not actuaries. We are industry owned. We have been involved in the various feasibility studies that have been going on for the last four and a half years. We were involved in one study with the Money Advice Service. We were also involved in the Association of British Insurers (ABI) prototype. We built a pension finder service (PFS) for that initiative.
Since then, Origo has split into two. The standards part is now called Criterion. They were involved in defining the data standards that would support what the dashboard would display. Criterion is still hoping to be involved in that.
We have supported the Department of Work and Pensions (DWP), and we have continued to invest in our PFS. We have scale tested it to 15 million citizens. We responded to the DWP consultation alongside the IFoA and numerous other organisations.
I should say that Origo has not yet been appointed to do anything officially on the pensions dashboard. Our aspirations are to be the PFS. We are an industry-owned non-profit organisation. We will have to go through a procurement process, but we have been involved for all these years. We have a project working group of numerous pension companies and administrators. We think that we are well positioned. We understand it from a technology point of view.
You might have heard of us from our pension transfer service. We do technology integration between pension back offices and financial advisers. We have the digital ID for financial advisers. We operate in that space.
We have had four pension ministers that have seen this since Steve Webb was involved in Pot Follows Member. Numerous other countries have had dashboards for a few years. Some of them did it with, compared to us, minimal fuss.
The Swedes, the Danes and the Dutch were talking at a conference last year. They gave us the benefit of their advice. They were slightly incredulous that we had not got around to it yet, that we were still in four and a half years of feasibility. “Why can’t you just get on with it? What are the objections?” Although when you dig deeper on the Swedish experience, it was hard. They started off in a non-compulsory environment, and it has taken them 8 years to secure a decent level of functionality and adoption.
We have not even started that journey yet. The good news is that the DWP has set the direction with respect to governance. Governance will be the responsibility of the Money and Pensions Service (MaPS), which was the amalgam of the Money Advice Service, the Pensions Advisory Service (TPAS) and Pension Wise. That organisation is just forming now. They were called the Single Financial Guidance Body for a while.
The consultation has said that state pension data must be there. The policy intent is clear. It is to find and view citizens’ pensions. They have said that it will start with a single non-commercial dashboard hosted by MaPS. It will evolve in time. Timetables for multiple commercial dashboards are not yet clear but Banks, pension companies, financial advisers and employers could define their own dashboards in the future.
The technical architecture is also defined. There are quite a lot of things still to be done. There is a lot of talk about digital ID. Will it be a private sector version of Verify, which you can use to look at your state pension entitlement and to do your tax return? There is no decision yet on what will appear on a dashboard.
Guy Opperman was quite positive about this at the Horizons Lifetime Savings and Investments Conference last week. He said that he will be quite hard on any pension provider or administrator that is not preparing actively for this. However, the reality is that they are behind where they said that they were going to be.
There were some recent DWP/MaPS roundtables. They were asking back-to-basics questions, which is frustrating when you have been through 4 years of feasibility. MaPS has appointed a principal, Chris Curry, from the Pensions Policy Institute. They are close to appointing an implementation director to start next month. We do not know who yet. Unfortunately, John Govett, their CEO, departed a couple of weeks ago. We know that they are filling some key roles of subject matter experts – one on procurement, one on data standards and one on technical architecture.
While all this forming and handover between DWP and MaPS is happening, there is an inevitable loss of momentum. There is no news yet on the Industry Delivery Group. You would hope that they would be industry grandees and subject matter experts that would come from industry bodies like the IFoA and from providers and administrators.
They have appointed consultants to work on a business case, roadmap and planning phase. This is a 12-week exercise that will take us through to September 2019. It was supposed to start a couple weeks ago. It is imminent.
The core components are the PFS, the digital ID hub and the governance register, which would oversee who could play in the ecosystem in terms of dashboard providers, pension providers, etc. There are no details of any invitation to tender for that process yet, but it is likely to go through a full European journal process which takes a few months.
The DWP has conceded that there will be no live dashboard in 2019, which is something Guy Opperman had promised us.
Are we obtaining clarity on the outstanding questions? Yes, there will be compulsion. It will be in the next Pensions Bill. Who will be the governance body? It will be MaPS. How will it be funded? It will be via a levy, details to follow.
Will state pension data be available? Yes, but we do not know precisely when. You can get your state pension entitlement through gov.uk already. This will be a slightly different iteration of that technology. It will be DWP doing it rather than Her Majesty’s Revenue and Customs. It will have to integrate with a dashboard architecture, which it does not currently.
How many dashboards will there be? It will be one public one initially, and at a timetable to be determined there will be several. What is the architecture? A single PFS plus digital ID. What is the digital ID and how will it be paid for? The verified circles talk about Level of Assurance. Level of Assurance 2 requires your passport, driving licence and a document checking service. It is quite tough to get through that Level of Assurance. You also need details of your credit history. Organisations like the Post Office, IDEMIA and Experian provide that checking service for the public sector. This is currently Verify. Five of them are interested in moving it into the private sector.
What will the dashboard display? This is a good topic for this evening’s debate. While there is no confirmation, pension companies cannot do the necessary preparation. They cannot get their IT teams working on the application program interfaces (APIs), the interfaces that would open the back-office admin systems to find and display the details. From the ABI project, the suggestion was to start off with: “What have I got and where is it?” So at least people can claim some of the £20 billion worth of lost pots. You can see what you have in terms of Defined Benefit (DB), Defined Contribution (DC) and the state pension. And, what are the DC pots worth? Current value? A recent valuation?
All the details of what the service level agreements are and how often these details would be refreshed would be finalised by the governance body.
Criterion did all the work around the data standards last time and would be well positioned to do that work. Going forward, they would reach consensus quickly if they worked with industry bodies like the IFoA.
Will there be a projection or illustration of retirement income? All the consumer research that has been done by those earlier feasibility studies says that that is what consumers want. They do not relate to pension pots, they relate to “What will I get in retirement each week, month or year?” And “How will it be calculated?” is a key question.
Will it be a big database? For information security purposes, you do not want a single base that can be hacked. The PFS does not store any data. It is a conduit. It is an authorisation engine that orchestrates the finds and then stores URLs or tokens in a central place that can then be checked. The data are stored with the pension companies or with the integration service provider (ISP) or administrator. With the ISP, if you are not going to get your back-end systems to talk directly to the PFS, you might put something in front of it. This is like a data warehouse or a single customer view.
There is an important question around the sharing of citizen data. Will they be able to share their dashboard with a financial adviser or with a trusted helper working at MaPS, as you can now over the phone with TPAS? Yes, that is there in the consultation. This PFS has an authorisation service. It works with a technical protocol called User Managed Access, which manages all the consents. This is GDPR compliant.
There are two levels of consent. There is one that fits within the DWP’s policy remit which says, “I give my consent for this PFS to find and display my pensions across the industry.”
The second level of consent is “I give the consent for this data to be held or persisted at a dashboard run by MaPS or a third party.” How does the pensions dashboard relate to Open Pensions? The pensions dashboard is an early iteration of Open Pensions. While there is a policy remit around the dashboard, there is not any regulation around Open Pensions. Some advice for the pensions industry, the data standards people, from Open Banking is even if the initial iteration of the dashboard is fairly simple, you should include a broader data set in the regulations. As the dashboard evolves to something richer in data content, you do not have to go back and change regulation. Open Banking has had a few teething issues with that.
The dashboard is an early iteration of Open Pensions. There have been APIs around for a long time. Origo and Criterion, as they are now, provide financial advisers with a set of data standards for looking at all the details of a contract or a pension or an Independent Savings Account (ISA) or another investment.
So, a proportion of the industry is used to opening its admin systems to a rich set of data.
There is no regulatory driver for Open Pensions yet. The difference between Open Pensions and Open Banking is different regulatory drivers. Open Banking is about opening up competition for the big banks. You must know where the banking accounts and details are.
You go into an Open Banking app, and you share your credentials with that app. You give them your digital ID for the banks and the credit card companies that you are involved in. In the world of pensions, if you have lost your pots in the first place, you are not going to be able to do that. You may have some pots for which you have never had online credentials, because they have sat there from employment many years ago. So, there is a finder element to the pensions dashboard that does not exist in the banking world.
Having said that, there are lots of crossovers in the road map to open finance. You can see all your information on the same app, be it banking, pensions or ISAs, etc.
A few words on how the pensions dashboard works. The consumer logs into the dashboard. The dashboard invokes the PFS. That goes off to one of the digital ID providers. In a federated ID world, you will have a choice of which digital ID provider you choose. You may already have one with Gov.UK Verify. Hopefully, they are portable into the private sector.
Once you have a digital ID, there is a Level of Assurance that the whole system knows who you are. Given that trust, the PFS can then orchestrate a search across, we think, 300-plus end points. Think of all the DC providers in the sector: the investment platforms, Self-invested Personal Pensions (SIPPs) specialists, occupational providers and administrators, and the in-house administered schemes.
It orchestrates the search. It returns a result back to the dashboard. It stores the URL at the PFS. The next time the dashboard comes in, if the pension is already found, it does not need to go back through the PFS, it goes direct to the pension company. It will check on the way back that the authorisation is still valid.
That is the technology; it is not terribly complicated. It is not the most complicated thing between us and the launch of the dashboard. Things like calculating retirement income, sorting out digital ID and 300-odd pension companies opening up their admin systems to these APIs are the critical path activities.
The Chairman: Following on from that then, why is the pensions dashboard more than a data collection exercise? What is it that needs to go on the pensions dashboard to make it useful and to achieve its objectives?
The DWP’s first objective is members being able to find their pension pots and view them. However, within their consultation response, they outlined some other things that dashboards can do “as a minimum.”
At the DWP, they talked about reconnecting people with their pensions. I think “Find and View” does that. They talk about trying to engage people with their pensions. Find and View can do that to an extent, too. They talk about people being in control of their pensions. I do not think that Find and View is going to achieve that. They talk about underpinning a sensible environment for obtaining financial advice. It is difficult to understand how Find and View is going to help underpin the ability to obtain financial advice in a standardised environment.
They talk about the simplifying and understanding of decumulation strategies. As an industry, those are still developing. Certainly, Find and View is not going to do that.
In 2018, the working party put together a research paper, which was published at the beginning of the year, outlining the difficulties around some of those things. Michael [Roe] spoke about an individual’s desire and need to have projected the benefits at retirement rather than, say, a current value, whether that is a DC pot or a deferred DB. That is not enabling individuals to take control of their retirement saving.
We will go over some of the issues that we are going to have to address, and this is an area that, from an actuarial perspective, I believe we can add most value. There is a great deal of technology that Michael [Roe] talked about, the communication features, etc. We are not the experts in those areas. We need to work with the experts in those areas and to make sure that we can contribute most effectively from an actuarial perspective.
The following two statements were in the DWP’s consultation response: “For too long pensions have been obscure, hard to understand and inaccessible to many. Pensions dashboards will change this.”
Two paragraphs later it states: “In the first phase of dashboard development, no additional regulations are needed to ensure consumers using dashboards are appropriately protected.”
There is a view that the industry could have done something on this sooner and been more proactive, but that is a pretty damning indictment of everything that has ever been provided from a pensions provider to an individual with a pensions pot somewhere, and I think that is slightly unfair. For me, there is a misconception here that we have the regulatory environment that enables us to provide all the things that a dashboard needs to do. That is something that we need to be shouting as a profession.
Certainly, when it comes to comparability of benefits in a consistent manner, the complexity of the UK pensions environment is such that it does not lend itself to easily enable people to understand what their retirement income will look like because of the different types of products and benefits that we have.
Giving a clear understanding of the risks involved with those benefits to people is something else that is crucial from a dashboard perspective, which we as actuaries should be able to support.
In terms of that phased functionality, I think – and Michael [Roe] said the same – it is important that we have some idea of what the long-term vision is, the multiple dashboards, so that when we put in place the framework on day one, and we are not going back to providers in 12 or 18 months to say “Yes, we forgot about this and maybe you need to give us that, and maybe you need to do this calculation over here, which underpins our ability to communicate through the dashboard.”
While it is clear that we are going to need phasing of functionality, some vision of the endgame needs to be taken account of right up front.
In terms of Find and View, the functionality is to find pensions, collect information and introduce the dashboard to members. Let us not get carried away with ourselves because what we do not want to do is try to boil the ocean and never deliver anything.
In Sweden, they built theirs piecemeal and tried to get different providers involved and target more and more people. That is something that is key.
A second version of the dashboard functionality could be talking more about the value of people’s benefits to them. That could be through click-throughs to individual providers, current fund values, options and guarantees, contingent benefits, fee structures – all the sorts of things that put an individual in a position whereby if they know what they want, they are now able to control their own retirement savings, as opposed to Find and View, which is just the highest level understanding of what they have.
Beyond that, whether it is still on a centralised dashboard or whether by this point we have multiple commercial entities putting their own front end on the dashboard, we have risk assessment, retirement planning, an interactive toolkit, choices about savings, decumulation options, etc.
If that is the case, where does the line get drawn between decisions that you should be making with your individual provider and what is held on the dashboard itself? I think that there are pros and cons and many different options there which perhaps we will go into later.
Within that context, we also have to think about some of the differences that we have within the pensions environment in the UK.
How do we deal with DB versus DC? The initial ABI project, of which I know Origo was a part, had some very basic information on it. It had a fund value for a DC benefit, and it had a pension amount for a DB.
What does that mean if I do not know the difference between DB and DC? It does not mean anything. How do we deal with simple differences between the two most common types of benefit within the UK? There is default decumulation for DB, but not for DC. Normal retirement age is an obvious feature for a DB scheme. Again, however, we have many schemes with many different retirement ages within it. That brings its own complications.
With DCs, we project to an arbitrary retirement age. How does that play? When you start bringing in things like state pension as well, does that bring in a third date or a fourth date for some people?
Regarding pension increases, on one side they are predetermined; on the other side, they are optional. I think ABI numbers are seven out of every eight pensions are non-increasing. Two out of every three have no contingent benefits. Is that a default that we want to use as part of projections within the pensions dashboard? Does that make them comparable or sensibly comparable against DBs, where these things are predetermined?
There are options and various factors that apply to individual DBs that are less relevant or not relevant at all in the DC space. We have talked about whether live fund values should be used for DC benefits. Is that the right thing to do or should there be a fixed point and you do not see anything again for another 12 months? There are again pros and cons for both of those options.
The DB side, obviously, is far less sensitive to market changes, albeit there are still some secondary impacts, such as employer covenants.
There are huge differences just between simple DB and simple DC benefits. Translating them for the individual is not necessarily that straightforward. We have talked through the issues with the various retirement ages. Do you introduce different benefits from different ages? Do you try to standardise them all, perhaps at state pension age, but then that may have little or no meaning to some people? How do you get around those issues?
Graphical representation of benefits would make sense to an individual so they can understand what is going on. We have had some relative success with this type of approach in terms of member response particularly in, for example, the civil service scheme where there are people with two or three – sometimes four – different retirement ages.
It may be possible to build a chart where you have a graphical representation of what your income would be at different ages. The question then comes from the individual, “Is that the only thing I could do?” Obviously, it is not. DC pots can come into payment at a different point. With DBs, you can take early or late retirement. You can defer your state pension.
The amount of data that is required to produce a simple chart is huge. If you want to make that interpretable by an individual and give them the option to consider different retirement profiles, then clearly you are exponentially adding to the required data. I am not going to suggest that that is what we should be putting on the dashboard. I am saying that if we are not careful, then we will get ourselves down a dark alley where there is no way out. We will have a product that is not helping individuals plan for their retirement.
Who sets the assumptions for the projections? That is an interesting point. The automatic response was we should use Statutory Money Purchase Illustrations (SMPIs). That does not seem unreasonable.
But there are flaws and inconsistencies with that which will potentially drive negative behaviour, if that is what we use to underpin the information on dashboards. We could have provider-set assumptions. That is achievable although it probably requires some form of tighter regulation around it than we have now. Like the SMPI approach, it can be consistent with a redefined SMPI, etc.
Communication is going to be interesting, particularly if the dashboard takes off. Ultimately, I think that if we get it right in the next decade, benefit statements will become a thing of the past because they are not adding value if people have access digitally. It may well be that some people prefer benefit statements. They can opt in to get paper copies, etc. Ultimately, digital access will probably be the way of the future.
Building on SMPI projections can capture the variety and complexity of different asset strategies, life-styling and expenses in which, from an industry perspective, different products have different profiles. That is legitimate. There is an argument for being able to include that and to cope with scheme-specific features within different products as well.
This does, however, require providers to do those projections. Whether or not we are using figures at a single point in time each year to populate a dashboard or getting live figures when an individual logs in means that providers might do the same as they do today and hold data on a system or you could be looking at having to provide live calculations on a minute-by-minute basis.
Those are the two extremes, perhaps, and it will be somewhere in between. But it does mean that if we are using provider-set assumptions or projections, then it becomes a more difficult proposition from a provider perspective.
If we do that, does it drive the wrong behaviour from providers? What flexibility do we allow? Does it drive an expected return competition? That cannot be good for anyone.
The alternative is dashboard-set assumptions, which reduces the provider’s work. That would be exactly the opposite: the same expected returns for similar assets. However, we do not get to take account of the complexities and vagaries of different products. It may be considered that that is less accurate, although accuracy may be the enemy of getting something that is understandable to the individual. That can lead to inconsistency with SMPIs as they currently stand. I suggest that we need to consider whether those are appropriate in a dashboard-led world or even just a world with a dashboard.
Here is a simple example of current SMPIs, which does depend on the information that is given to individual providers. Take a 25-year-old, earning £25,000 a year, an estimated state retirement age of 68 and contributing 10% of salary. There are projections out there where that 10% of the £25,000 salary is £2,500 a year from now until age 68. There are others where it is linked to inflation; and others where it is linked to some form of earnings assumption.
How do we deal with different risk profiles? How do we communicate these things to individuals?
Investment risk, longevity risk, inflation, covenant, data, and personal health are all relevant to different people and different products. There is no clear way to compare one product to another.
In many of these areas, there is no way that it makes sense to the individual. The simple suggestion is to try to develop something similar to the energy efficiency scale. That would be some simple form of expressing this to individuals; something in traffic light or rainbow style where the benefits and the funds that they have stack up.
Some of the countries that have a dashboard already have not had to face the complexities that we have in the UK. Some of the guarantees and options that we have here are difficult to deal with: guaranteed annuity rates, scheme-specific ones, protected pension ages, etc. We have different factors for different schemes. Collective Defined Contribution might come along.
There are many complexities across the board. How do we make sure that the individual is not ending up with a parsnip when they wanted a pear?
There is a need for regulatory change to support pensions dashboards. If we do not get it, we risk getting things wrong from an individual perspective.
This consistency across different types of pension communications is key to the success of the dashboard. If we are not clear about future functionality, and where it might take us, then we might not have the building blocks to deliver the things that we need to deliver.
My personal view is that we need to be quite vocal about it. Certainly, I am trying to be vocal in the various roundtables and working groups in which I am involved. I hope that we will be heard, and we will be well represented across them and make sure that the dashboard is the success that it should be. Difficult as it is, I believe this is a once-in-a-generation opportunity to engage people with the concept of retirement savings. If we get it wrong, then it will be another generation before we get another opportunity. As we all know, trust in the financial services sector is hard to build and easy to destroy.
Does anybody have anything that they would like to add or discuss around the dashboard?
Mr J. Taylor, F.F.A.: Is there any vision about the change in customer behaviour? Do we want people to be engaging more, saving more, and changing their pots to those with the lowest charges?
We are mindful that there are potential pitfalls here. For example, will people be attracted to a large DB transfer more so than the income that it is going to provide. Indeed, I think that there has been research in the United States that people who engage too frequently with their savings become risk-averse because they see the fluctuations and move to recklessly conservative assets.
What do we think is good customer behaviour that we are trying to encourage and what pitfalls are we trying to avoid?
Mr Roe: I think that there is a general level of excitement that the dashboard will be engaging for members and citizens to see all their pensions in one place. Consumer research has revealed a few reactions. Some people will be scared by how little they are saving and how poor they might be in retirement. That is a common reaction.
Some people may be tempted to cash money. Just seeing “What have I got and where is it?” begs further questions: “Should I save more?” and “Should I consolidate?” The Australians have transfer buttons on their dashboard and can consolidate straightaway.
From a technology point of view, you can see joined-up journeys. You can go from a dashboard to a consolidation journey, which may require opening an account with a SIPP provider. You can see all these things joined up.
I think that is getting ahead of ourselves. I think we have a few years just making the dashboard a success. There will be an implementation phase, like auto enrolment; it will take 3 or 4 years to happen. Many consumer journeys in the early years will be frustrating because they will see only one or two of their pension pots when they may be expecting to see five or six.
Another scenario is the stock market going down, so it is showing expected retirement income going down. It is making you think that you must work another year or – it has gone down again – you must work another 2 years. There are all sorts of real-time engagement like that which could be powerful – positively and negatively.
The Chairman: Different stakeholders are going to have different objectives. I alluded to some. The DWP’s stated aim is the minimum of a successful dashboard. I think there is one that is missing and is going to be an objective of government, which is that individuals should save more for retirement, but not so much that the tax revenue falls. There is a balance there.
From a provider perspective, a consolidation button would be great. I am sure many people would like to see that as an opportunity, but it cannot be an opportunity for everyone. Certainly, negative behaviour like taking DB transfers is something that we want to avoid.
One of the main arguments for not having real-time fund values is the constant decision-making and the potential negative impact of that. I think that there is a strong argument not to have real-time fund values. But what is the right time period?
It is about getting the balance right. What is positive behaviour? To me, it is going back to the point of individuals engaging with the concept of retirement saving; getting them to trust the fact that these numbers make sense.
I was at an event last week where we were talking about trust in the financial services industry. Part of it is not so much the actions and bad experiences in the industry. It is the concept of what it is that we are trying to get people to do, which is to give a predetermined amount of money to someone they do not know for an undetermined amount of time to do something that they do not understand, at the end of which they may or may not get some money back, which they might not need. That is what we are asking people to do when it comes to saving into a DC pension.
We all know that is not true. But that is how it feels to some people. One of the things that will be massively valuable will be an understanding of what it is that their investment is doing. What is it? Is it an infrastructure project? Environmental, Social and Governance investment can play a huge part in engaging people with pensions alongside a toolkit that the dashboard can deliver. That is the first step to good behaviour: engaging people with the concept that retirement saving is good. Where we go from there, I think is up to us as an industry.
Mr C. S. Smith, F.F.A.: Following what you have just said, I was just wondering what part of society are the DWP targeting? I expect over 80% of the population will never look at this. They are possibly those people who are most in need of seeing this. Have other countries that have done this seen any change in the behaviour of consumers regarding this information?
Mr Roe: It is designed to have universal application so every pension company will be compelled to provide the data. You can imagine that high net worth individuals who can afford financial advice tend to have consolidated into SIPP-type arrangements. The rest of the population will benefit from this.
Another sector of the population that will benefit are those who have lost touch with their pension pots. The ABI put a figure of £20 billion on the amount of lost pension pots. The sooner we get the PFS doing some finding, the sooner some of those older people in society will be reunited with pension pots. That must be a great social purpose.
In other European countries, I know that there has been a lot of research on consumer behaviour, but I cannot comment on specific findings.
The Chairman: There is evidence in some countries that saving has increased. The Israeli dashboard is commercially focused. That has led to more movement between the providers.
Those who will benefit from this most depend on how it is designed and how it is marketed. The behavioural economics side of the design of this is crucial.
If we want people to engage with it, we must make it talk to them. That might mean a different thing to a 20-year-old and a 50-year-old and a 75-year-old. I think it is not so much who are we targeting, but how we enable it to become a universal tool.
Digital coverage across the pre-retirement population is in a market which is almost saturated. You are, even in the 60-plus ages into well above 90% digital penetration. People have the skills and the tools to engage with a digital product like the dashboard. But what does it say to them when they get there? What is important to a 20-year-old is not important to a 60-year-old, and vice versa?
How do we work collectively with a broad skillset to make sure that the dashboard delivers the right thing? I suspect that is something that our profession can contribute to, but probably should not drive.
Mr A. H. Watson, F.F.A.: What are the thoughts of the working party on how we communicate the risks of DB pensions and that you might not get exactly what it says, and you could end up in the Pensions Protection Fund (PPF)?
The Chairman: This goes back to the crude illustration of how we might demonstrate risk. Covenant risk is important in the DB space. Some form of standardised assessment of that is going to have to be a part of assessing DBs. If I am an individual and I am taking appropriate action with my benefits, that is information that I can get hold of that can influence my decision-making.
At what point does it become scaremongering? There is something about making sure that the information is in a complete but standardised format to support taking proper regulated financial advice.
“Robo-advice” is a bad combination of words in certain places. I am not saying that I am an advocate of it. I am not saying that I am not. The dashboard can provide a standardised data set with which to do something.
That must include covenant, sponsoring employers and risk of PPF, and what that then might mean. The information that we are talking about here is not Level 1. We are three or four or five layers through by the time we are talking about that sort of thing. The majority will not be interested in digging that deep. If we do not give them Level 1 information, it will never get to level 5. If we do not give them Level 1 information, they will not be engaged at all.
It must be included in the risk assessment. How you do it, I do not have the answers yet. That is where we are, and we need to work it through.
Mr J. A. Bielski, F.F.A.: In terms of our comparison between DB and DC, are you including in the dashboard the transfer value? At least you would then have the current value. You would have a comparable. That might not be what consumers are interested in, but it might be something that at least gets you to Level 1.
Another question is about non-pension product assets. Many people talk about property as being their pension. They talk about ISAs or they just have savings. Without being able to see that, or input that, is it really a pension dashboard at all for a lot of people?
The Chairman: I think the question there is: is it a retirement dashboard? If we take the first question, I think that if I were to say to my father-in-law, who is smart and coming up to retirement, “There is your DB transfer value from all your years over here, and here is your DC pot for the past 10 years over here,” he is going to say “So what?”
He needs to be able to access that information. I agree. I think that is probably after he needs to be able to see what it is he is likely to get if leaves it where it is. The reason I say that is because those will look like gargantuan sums of money to many people. The only time they have ever talked about sums of money of that size will be when they have bought a house. And that is even if they have bought a house.
I am committed from a philosophical point of view to giving people information about their benefits and options. I think that is crucial. We must think from a behavioural perspective what is right for the individual. I am clear that that is benefits first, with fund values and current transfer values a layer down. It probably must go a long with a degree of description and caveat that you do not want on page one because it may disengage.
In terms of it being a retirement planning tool, I refer to the multiple dashboards option where commercial entities have their own front end. I agree with you, that needs to be a key input. Personally, I would not see that as something you have on a centralised tool.
If you were an Independent Financial Advisor trying to use this information for: “Login here to our version of the dashboard and we can help deliver you some global advice,” or whatever else it might be, then, yes, you are going to need to have that ability to enter other assets. That is such a huge proportion of some individuals’ retirement savings. I think that is why this tool in the dashboard is so crucial.
People do invest their retirement savings in cash, let alone property, because they can touch it, feel it, they trust it; they know what it is. They do not understand what a pension is a lot of the time.
While education is not the only solution to that problem, I think the dashboard promotes awareness. It promotes a closeness to your benefits. You can see them. You can touch them daily like you can your current account.
If we do not get ourselves into a position where we are at the same level as other industries, people will put their money elsewhere because again they do not understand the benefits of saving within a pension.
Ultimately, I can see a scenario where Open Banking, Open Pensions and Open Tax Returns, and so on, talk to each other. If you want to move money from one account to another and into your DC fund, and you want a tax rebate and then you need it back at some point, that is possible. Technologically, that is possible. If you want to engage people in the context of retirement saving, make it easy.
I would say that we must include those things, but probably not at the beginning.
Mr Roe: We are already at the stage where you can see all your money in certain applications (apps). You can see savings and investments, pensions, loans and mortgages. Behind the scenes, there is not much pension company connectivity.
In practice, consumers can go to those apps and say, “Here are my details.” That would involve a point-to-point connection. In a lot of cases, it is just a letter of authority going on behind the scenes. The advantage of a PFS and a dashboard is it is there in the regulations. The pension companies must make the investment to connect to it.
If it is an investment platform and it is advised to those platforms that they want to make the client details available to direct-to-customer apps, that is a whole landscape which is at an early stage, I would say.
Having said that, Open Banking is making inroads. The dashboard will be the first iteration of Open Pensions. The tax incentivised savings association, for example, has an initiative called Open Savings and Investments. Historically, they were the ISA trade organisation. Whether these things all merge into Open Finance at some point in the future is aspirational at this stage.
Ms C. M. Stewart Roper, F.F.A.: I agree that benefits are what are important to people. But how do you illustrate DC benefits in a pension freedoms world? And, when you start doing so, how do you get people’s personal longevity risks into the pensions dashboard?
The Chairman: These are the difficult questions that we have. How do we do these things? Your assumption setting probably must start with something that is standardised.
Should we let 100% accuracy get in the way of delivering something usable? There is a balance there. Again, I suggest that some of the things that you are talking about come in at later stages in the commercial offerings, whereby you are getting added value from interacting with an organisation that is doing something specifically for you rather than getting something from a generic dashboard. But that generic centralised dashboard must be a starting point. That is the difficulty that we have. How do you translate DC into something meaningful?
Decumulation to me is a bigger issue because what does drawdown mean? What does the profile of it look like for any one individual based on their circumstances? Annuities will become a thing of the past, if we move to a world where the decumulation market is more mature and has more options in it. What does that look like? When does that happen?
We do not have answers to those questions. That is why this is hard. That is why, when we ask if a regulatory environment exists to get something meaningful from an individual on a dashboard, I do not think it does. That is why I think the actuarial profession needs to be well represented when we are discussing these matters.
Ms C. L. Kingston, F.F.A.: Can you talk about the move from a generic dashboard to a commercial dashboard or a multiple commercial dashboard? Who would fund it? Would it be by subscriber? What is the experience of other countries?
You said that the dashboard would be paid for by a levy. How does that then translate into a commercial model?
The Chairman: That is an interesting question about the levy that will pay for the dashboard. Ultimately, wherever it comes from, it is going to come out of the saver’s pocket.
How do we go from a centralised, single dashboard to multiple, commercial dashboards? I think that is a hard journey. I do not know how quickly that will happen. Michael [Roe] might have a view on that. I think that it is going to be up to individual commercial entities to decide whether there is benefit in doing it.
We flippantly mentioned a consolidator button might feel like a good value-add from an individual provider perspective. It is if people are using their consolidator button rather than someone else’s. You alluded to that as being something which has led to action in Australia. It is hard to know how quickly and how realistic it is for organisations to think of a commercial dashboard and what that offering will entail.
I have some ideas about what you could do in terms of a pensions smart meter or a pensions satnav. Where are you? Where do you want to go? How do you get there? But what does that look like in the long term? We must make sure that we get the first few stages right before we carry on into whatever the commercial dashboard landscape looks like.
Mr Roe: I have a few comments on that. When the Treasury was running the ABI project, the Treasury was very keen on FinTech innovation and competition. They wanted to see many dashboards from the outset. This is one of the debates that goes backwards and forwards. The DWP have gone back to where the Money Advice Service was 3 or 4 years ago, which is: “Let us start with a single dashboard. It will be easier to control. It will be safe for the consumer and data protection. Let us worry about innovation and competition at a later stage.”
Most other countries have gone for the single state-backed dashboard at this stage. Many people in the industry would like to have their own dashboard. They want to differentiate themselves and do creative things. There are many FinTech companies who want to take a dashboard and plug it into modelling tools, retirement income, cash flow, risk management and so on. And why not? Why not open the door to innovation further down the track?
Those are the debates that are going on. If you want to enter the sector as a dashboard provider and you will gain from it as opposed to a pensions administrator who will not be providing a dashboard, and just have the IT cost of complying with the regulations, then those dashboard providers should probably pay you to be in the ecosystem.
Just one other thought on the cost of advice. There has been speculation that much of the groundwork that financial advisers have to do at the moment is the basic legwork of finding a client’s pension pots before they can offer any advice. That could be at least a third of the effort. If the adviser can get delegated access to a citizen’s dashboard, then a lot of the cost of the initial fact-finding will be taken out of the advice process. It is hoped that would be passed on to clients.
The Chairman: Any more thoughts or questions?
Mr Taylor: If I may reflect on the pensions freedom question, we have done a fair amount of modelling, looking at the differences between the income yield you might get from a sensible drawdown plan at a reasonable level of sustainability versus an annuity. For somebody in their mid-60s, it is not a huge difference in the income you might get. Your point, Andrew [Lowe], would suggest you can give a Level 1 sense of benefits. It might be 3.5%, that sort of ballpark; that would be meaningful without having to get into the annuity versus drawdown consideration. Maybe that is a Level 2 or Level 3 consideration. That was the point I wanted to make on the dashboard.
Mr Watson: Is the working party going to continue? We need this working party to continue. We have Level 1 but need to start delving into Levels 2, 3, 4 and 5.
The Chairman: Our first difficulty is getting this discussion on the table with certain people. Michael [Roe] alluded to this during his opening. At the roundtables, probably 20% of the people in the room understood that there is a problem here that needs to be addressed.
The working party needs to be involved in the Industry Delivery Group. It needs to be coming up with a framework of comparability for these different complexities. That starts with the simple things like how to deal with different retirement ages. How do we deal with DB versus DC at the very highest level?
As we go through the evolution of dashboard functionality, there are ever more questions that we need to answer. To get to the bottom of that and do it properly, we need more people with the ability to break down the problem and solve it before this will work. It is not impossible, but it is hard.
Mr Roe: It would be brilliant if the actuarial profession could step in and give us some firm and early advice on how expected retirement income could be calculated for the first iteration of the dashboard, which maybe will not be live this year but it will be in Beta format. Technologists, administrators and civil servants are not going to make those decisions. We need the actuarial profession to tell the rest of the organisations what to do.
The Chairman: We have reached a natural conclusion. I do believe that this is important. But if we get it wrong, it is going to set us back tremendously. It is quite easy to dismiss it as a data gathering or an administration exercise combined with the technology that needs to underpin it, but it is not.
We are working with companies like Origo to deliver a successful dashboard.
Open access
