Two world-leading experts on pension dashboards explain how they might affect engagement in the UK, and the importance of good data, robust controls, inclusivity, and managing expectations.
Voiceover: The Thinking Forward Podcast brought to you by Standard Life.
Sangita Chawla: Welcome to our Thinking Forward Thought Leadership Podcast series. Here we discuss trends and developments impacting the UK pensions industry in order to understand how more people in the UK can achieve financial security now and in later life.
I'm Sangita Chawla, Chief Marketing Officer at Standard Life. Today, we're here to discuss how the UK dashboards can learn from what's happened in both Sweden and in Denmark. I'm joined by two guest speakers today, Anders Lundström, CEO of minPension in Sweden, who's responsible for the Swedish dashboards, and Michael Rasch, who's head of the Danish dashboards. Michael, Anders, welcome to you both.
Now, as we make a start, I'd like to ask Michael the first question. We've heard a lot about the Swedish dashboard experience earlier in the year. Michael, it'd be really great to get an overview from you of how the Danish dashboard came to there. What were the main features that you thought about when you were building it? How many people use it? What's been the key things that you've learnt along the way? If I could invite you, Michael, to comment.
Michael Rasch: Yes. It started out in 1999, where a large provider, the largest Danish provider, took the lead and built with a few other providers the website. There were no need for legislation. That also meant that it took some time before the whole industry-- They didn't join until 2007. Then, from there on, it has just been an ongoing development, and more and more users have been joining the website and using the website.
It's a private association, so there were no need for legislation. For now, a lot of things have been put into the dashboard over the years. One of the main features that is used quite a bit is the users' ability to send all their data themselves from the website to a provider or a bank or a counsellor, or others.
In the later years, you can also, on the dashboard, get an overview of your household. What is the household? If you get the acceptance from your spouse, you can get both you and your spouse's information shown on a website to get an overview of how the household income will be when both of you retire.
On the website, pension forecasts are shown from the earliest retirement age, that is around 60 for Danes and until 70. You can see what you can expect to get in your retirement at different retirement ages, from 60 to 70.
About the usage, it started out very early. In the early years, it started out very slowly. Only very few used the website because only a few providers were on the website and delivered data to the website. It didn't really give much value to the user. Today, we are a small population of 5.8 billion people, but on a yearly basis, we have 1.6 million users on the website. They log on close to 6 million times a year.
We are very glad for the usage on the website now, but it has taken many, many years to get there. Don't have your hopes very high regarding the UK dashboard in 2023 that a lot of users will use the website or your dashboard from the start.
Sangita Chawla: Thank you, Michael. You mentioned it was privately funded. Has there been involvement from the government at all in this?
Michael Rasch: Yes, there has been a close contact with the government the whole time. We have a board of directors of 11 people, but we also have representatives from two ministries attending the board of directors meetings. We have a close contact with the government through the ministries. That worked out pretty good.
Sangita Chawla: Thank you. If I move to you, Anders, and listening to Michael there, how would you compare and contrast the Swedish experience, so thinking around the government involvement, but also around just the sort of roadmap from start to where we are today?
Anders Lundström: Thank you, Sangita. I must say that I'm impressed by the Danish, and we work together as more like a pension industry together. In Sweden, we had different problems than we had in the States. It was the driver of getting up a pension dashboard. That was due to the new pension system that we got in the late 1990s, and there was a huge demand or request for more information.
It was a little bit of a sticking card game between the state and the private pension providers. It took-- I would say, just like Michael said, it takes a long time, and you need to be very thorough in your work.
I think that we today have a very good cooperation between the state and the private pension providers. If you compare it to [unintelligible 00:05:36], we are fully a private-public partnership, and the state takes 50% of operational costs, and they do have a lot to say about the different priorities, too.
I would say that when we started five years later than [unintelligible 00:05:52], I would say that we had about the same starting point and the same way to approach our users.
If you look at the different stakeholder perspectives, I would say that we have a much stronger stakeholder perspective from the state than we have in [unintelligible 00:06:11] in Denmark. I would say that's the big difference in between here.
You can see that the Danes have managed to cooperate with the data sharing. We haven't cooperated to do that, and we haven't got any solution yet. I think that's the main difference between the different countries here.
Sangita Chawla: Yes, and I think that's one of the things that's very tricky for all countries to actually reconcile with, and that is around the use of data and how that's actually protected and therefore then shared across, so really good point on that.
If I could push you on the user experience and how you actually went about designing the dashboard, was it something that you did with a team of experts sitting in an office, or how did you get to the point of creating the user experience?
Anders Lundström: First of all, this is one part where the state was very strong because we spent a lot of time to get more and more information and get more details and more specialist functionalities for the more experienced user. We didn't fully look at the full user group, and we could see that we had developed something that was more for the expert user or the well-informed users, but we needed to find a way to have all the user segments represented on the website.
We had just one website that was the solution for everybody, but what we started to do in 2015 was to start with user segmentation. Actually, we had to have a totally different approach to attract these users because we needed to be much simpler, the simplicity on the dashboard and also get information that was more what they asked for.
If you're about 25 years old, you don't want to have a pre-action or a forecast, you're more interested in having a checkbox, what should I think about, do I need to do something now or can I come back in two or three years or whatever.
We started to do a lot of user testing, and we had about 10 different segments that we used and tried to see what kind of functionality and data did attract the different users of the different segments, and then we started to look at the Swedish population, and together with the Swedish Pension Agency, we made a segmentation of the Swedish population too, and there we could see that we didn't reach people with low interest or low knowledge, and they couldn't make the right choices. They didn't make any well-informed choices from that information that we could see on the dashboard at that time.
We started to do a lot of testing, both user testing, qualitative testing, and we also built a separate beta site running on production data. We invited users to test our new beta site and get their feedback on the dashboard. That way, we put about three years of work into that. That is the dashboard that we are using today, but you need to have a kind of structured way to measure different data and how you can attract different parts of users, too.
It's quite a long learning, it's quite a steep learning curve to try to do that, and you need to be quite brave to do the different experimentation with the dashboard, too. It's not an easy task.
Sangita Chawla: Not an easy task at all, and I know when we try and do these things here at Standard Life, getting user testing, it's always tricky to hold your nerves, isn't it, on what the insights show you and what you should push forward with? I'm curious, Michael, did you have a similar experience in Denmark? Did you approach in the same sort of way, or was it quite different?
Michael Rasch: A bit different. I'm very impressed by the work they have done in Sweden, and they have gone very far, and come very far. We have had five different ways of presented data. The website has changed five times, and we have gone from the more descriptive sites to a more layered information site.
The last, I would say, seven, eight years, every time we have had a larger implementation of new functionality, we have had user testing and also several user testing sessions throughout the whole development process. That has worked out very good and we haven't-- let's say, the household income part of the website wouldn't be where it is now if it hadn't been for thoroughly user testing prior to implementation.
We are no near the level of user testing and user analysis that they are and what they do in our neighbouring country, Sweden. I follow their work very closely, and we are having an ongoing focus on the value of user testing.
Sangita Chawla: Right. One of the things that we talk about here in the UK, as we're thinking about dashboards, is this aspiration, really, that we think when dashboards come about, that people will become more engaged with their pension and as a result of that, may take differences in behaviour, they may do something around their pensions, or at least they may find old pensions.
A recent paper that came out by the PPI was that there's an estimated 26.6 billion in lost pensions, which is almost 40% up from the figure in 2018. We sit here and think, well, if we get dashboards working well, then hopefully people will find their pensions and then be able to interact and engage and obviously take action. Have you experienced a change in behaviour, or seen people do different things with dashboards now that it's up and running? Perhaps, Michael, if I could start with you?
Michael Rasch: Yes, we haven't seen a significant change in the behaviour. For many Danes, it's still a difficult topic, and they don't have a huge insight into their pensions. What we can see that especially when you're sending information, let's say, to your bank or your pension provider prior to a meeting, through the information, they can see if you have, for example, dormant accounts.
That's something we can see a huge change in that a lot of dormant pensions accounts at different providers are transferred to your current and the user's current pension provider.
That has been very useful and is an ongoing process because there's still people changing jobs.
In Denmark, we don't have the same issues as you do in the UK regarding lost pensions, because all pensions are connected to your national identification number. When you log on, either on the website or when you change address, the pension provider will get information about your new address if it's in Denmark. You will not, in a very, very little scale, not be able to find your pension. That's a great difference in Denmark regarding to the UK.
Sangita Chawla: Anders, how does the experience compared in Sweden then and are people actually doing things differently?
Anders Lundström: Yes. First of all, it's not a quick fix, it takes some time to get the change in behaviour. I must say that we can see that people all different pages and all different knowledge groups or different user segments, they are using the dashboard today, and people do have much better information for doing their well-informed choices on their future pension.
What we specifically can see at the moment when we started to do more specific functionality for different user groups, like when you're very close to retirement, we built a retirement planner. One of the things that we can see at the moment is that people seem to do different choices. We can see that people stay in work longer and they start with part retirement, so they start to work about 80% and have one day off for something like that as retired, and then they continue to work and have much more retirement planning actually.
People doesn't just go into their HR department and have a coffee and a cake and says, "Well, I'm retired tomorrow."They have much longer working life and stays in work longer, and they do much more of planning with their retirement, too, and pensions. I can say that's one of the biggest changes we have seen, but we can also see a change in behaviour, with people more opting for having occupational pensions than before because we have started to ask their employers about occupational pension. We can see that the people in low ranges, they more often have some occupational pension or some private savings on the side.
We hope that we can present a report here next year during the changed behaviour. Actually, we have a big study going on at the moment, so I hope I can give you some more facts later on next year.
Sangita Chawla: Absolutely fascinating. I guess what you're both saying is that the UK needs to be much more realistic about its expectations of what will happen when dashboards get launched next year, and we have to bear with the public on how much engagement it actually drives if I read that right from both of you.
Michael Rasch: I agree. I think one very important element in this is the dashboard isn't better than the data that it's put into it. I don't know of the rollout plan in the UK, but if only you can see a smaller number of your pension schemes from the beginning, you shouldn't have high expectations on both the usage and the people's knowledge on pensions getting better.
When you are reaching, let's say, 80% or 90% of the pension providers in the UK deliver data and data of a certain quality to the site, it's at that point you really can start to better the knowledge of their-- and the dashboard can help the users getting higher knowledge on their pensions. If you only have, let's say, 15% of your pension shown on the dashboard, it doesn't have that value that it will when it's fully built.
Sangita Chawla: Agree with that. Anders, did you want to say anything on that? Sorry.
Anders Lundström: Yes, please. I think that the UK pension dashboard will have a lot of challenges, and I know from my own experience that it's quite a tough job to start up dashboard, especially when, like Michael said, that you have expectations on data. In the beginning, you won't cover that much of the pension industry and market, so a user won't get the full information. You will get some kind of a catch-22 situation in the beginning. If you don't have any data, you don't get any users, and if you don't have any users or that many users, it's not interesting to deliver data to the dashboard too.
I understand that you have a legislation there, but it's still a long way to go and to populate with pension data into the dashboard. I would say that media, for our sake, when we started up in 2004, they set up different expectations on what you could do with the dashboard, and it would solve all the problems within the pension industry, but we could only give part of the functionality to do that.
The expectations were so high, so people were so disappointed at the beginning. They start saying that, "Well, it's no use with this dashboard, doesn't solve anything for me." That you need to be prepared, but it will take a few years before you get there. In our case, I would say, it took about five years before we had an attractive product or attractive pension dashboard from users.
You need to set the expectations from the user at the beginning, and you have to work quite closely with the media to set the right coverage and also tell the story about the dashboard and have some communication on how a dashboard will be populated with more data and what use you can have of a dashboard.
Actually, we did set up quick functionality after the beginning for people to add in their own data, so they could have all of the information in one place. That was some workaround that we survived with. You need to be prepared to do different changes and prepared to change from the beginning. I will be very happy to look at the UK dashboard from a site actually.
Sangita Chawla: I'll be glad to share it with you when it comes, Anders. Both of you have just actually mentioned data there, and that's something that we constantly think around, sort of the quality of data that goes into dashboards and how you make sure there's quality assurance going on. Because, of course, unless the data is of a certain quality, then the information to the person is going to be limited.
Anders, you and both Michael mentioned data there, do you have any words of wisdom for us in the UK? Start with you, Michael, on this one.
Michael Rasch: Yes, I think it's something you maybe don't consider that important, but it's so crucial. What we have looked into from our Danish perspective is that the data quality in many ways is pretty high, but we have pension forecast. Even though we have been running the website for more than 20 years, there are still pension products and providers that have some challenges delivering all the needed pension projections to our website.
It's an ongoing focus, and especially in the UK situation where you are starting the website, and many have a view on what they expect the quality should be. They expect the information to be correct in every way. That is, I would say, you might have to have some certain way of-- the pension providers have to deliver data of a certain data quality before entering to the website because poor data quality can, especially in the beginning, be very harmful for the public belief in the dashboard, so that's important.
Then also agreeing on what should be the mandatory data delivered from providers to the website and also information onto the users, information to the users. If there's certain information that is lacking from one provider, it makes it quite difficult to show that these informations are not shown on the website. That's also a task for developing the dashboard.
Sangita Chawla: Thank you, Michael. Anders?
Anders Lundström: I'm a little bit boring, but standards, standards, standards, standards, that's what you need to do, work. Takes a long time to do that, and I think we are in our fourth version of information standard, but what we started to do now is that we have built in some control points in the data delivery, so we have to do some kind of more like self-testing before we can send the data.
As Michael mentioned, a lot of more data points will be mandatory, too. I think that the main thing here is that you need to find a good way to engage the pension providers in the data delivery, and we actually created some tools for more like self-help to the pension providers. We can see different changes in the data and how will be published on the dashboard.
Actually, we are also doing some control points ourself for comparing data. If we have data before or if it's new data, we do some quality checks ourself. You need to be working quite closely with the pension providers to get the right set of quality. It's not a funny work, but it has to be done, and you need to engage the pension providers.
Sangita Chawla: Thank you. Just moving on to a slightly different topic. The dashboard is great, obviously, for people who are digitally engaged, and we know that not everybody is, whether they don't want to be or whether they don't feel confident with interacting online. How did you think about reaching those people, and did you have many people that fell into those categories? Perhaps, I'll start with you, Anders, on that one.
Anders Lundström: Thanks. I would say that we have about 600,000 people in Sweden today in different ages who doesn't want to be engaged in digital tools or on the internet. Compare that with 6.4 people who is in work, so it's like 10% then. What we can see is that we try to find new ways to reach those guys or people and actually working together with the Swedish Pension Agency.
We have set up a special service for their customer support that people can dial in and do some identification with a pin code and some extra service there, and then they can get help by phone, and the customer service can try to do the simulations together by phone, and we can help them there.
We also have a pension agency. We have more like mobile customer service who's standing out in shopping malls, things like that, and do a hands-on help to people who get in there. Actually, the state provides with some tools for us, or we work together with them to reach for people who doesn't want to use digital services.
As I mentioned before, we need to be much simpler to get people to use a service, especially for people who just want to avoid the different services on the internet. I think you have to work quite hard on that, and you need to do a lot of work just to get them there.
Sangita Chawla: Thanks, Anders. Was your experience in Denmark the same, Michael?
Michael Rasch: A little different. We had some help from the state. The state, I think, four or five years ago said that all Danes should communicate digitally with the different state tax authority, all the public authorities. It should be a digital communication. You had to apply not to have a digital communication. Many of the Danes were somewhat forced over to using the digital platforms. That has helped us quite a bit.
There are still a small number that are not capable of using the digital solutions like the dashboard, but we don't have any specific services regarding them. All the different providers send information to their customers in a paper form on a yearly basis. They get the informations on their pensions in a paper form yearly if they're not digitally. What we can see of the numbers from the state, it's a group that gets smaller and most of them are on the high side of 70 years old.
Sangita Chawla: That's actually really interesting that it was actually mandated to be digitally-led. Fascinating. I'm conscious of the time that we've taken up with you, gentlemen. I'm going to wrap up with one final question to you both, which is what are you most excited about to see in your own country with dashboard development and what is the one thing that you give us a lesson on for the UK that we'd need to think around? I'll start with you, Michael, first.
Michael Rasch: What is going to happen in the years to come? At the moment, we are also looking into having a service where you can see how does the pension development in real time value develops over a period of 30 years. You give information on your pensions on what is purchasing power of your pension throughout the whole period. That'll be launched next years.
Beyond that, we are looking into the whole state pension. We had to do a much better information on the state pensions on our websites. That should be very helpful for the future people that retires, and where you can also have a good idea on what if I work at the same time as I get my state old age pension.
That's also something that I'll think that will be more common regarding your other pensions that you start with your pensions and some of your pensions, and then you only work part-time. I think that's something that our websites need to support within a number of years.
Anders Lundström: For our sake, we're looking into a flexible retirement age as the number one challenge to get people to understand that we have to work a little bit longer than before. We also have consumer information on policy transfer that is coming up that we hope to be presenting in the first half year next year.
Of course, we always have a data sharing that is popping up and down. I think that we are moving close to find a better solution to handle that. Regarding the UK, you will have a big challenge. I think we had a lot of problems, and we can see that the maturity or internet maturity of different pension providers were quite low at the time, but you need to work a lot with the data quality in the beginning.
I think that you should start small, try to keep the expectations of a dashboard quite low, but think big and try to be very iterative in your development of a dashboard. I'm quite sure it will change from what you believe today and will be in two or three years' time, so good luck.
Sangita Chawla: Thank you, Anders. Any final words of wisdom from you, Michael, for us, about the UK?
Michael Rasch: I would say the part that you have many different dashboards doesn't make your whole setup easier. One is data quality, but also, it's also an issue if you present data in different ways, and you have 20 different dashboards. I think there's also something that you should look into. If, who's responsible for the presentation of the data, what is, if data is presented wrongly, it's delivered correctly, it's the right quality?
What if data in one dashboard is presented wrongly and gives a user a wrong impression of both the size of the pensions and when it gets paid out? That is something I would recommend you, too, to have an open eye on in a central way that this will probably happen, and could there be legal issues regarding this? I don't know if there could be some liability or anything but something that you have had thoughts on how you should handle.
Sangita Chawla: Thank you, gentlemen. That was brilliant. We look forward to sharing more of our dashboard development next year and in the years beyond to actually share what's worked, what hasn't worked with you, and to hear more from you.
That brings us to the end of this episode. I'd like to say a warm thank you to both Anders and Michael for joining us today and what the UK can learn from dashboards in both Sweden and Denmark. Thank you, Anders, and thank you, Michael.
Michael Rasch: Thank you, Sangita. Thanks for having us.
Anders Lundström: Thank you.
Sangita Chawla: Thank you very much, and we hope our discussion was beneficial to you, our listeners. Thank you for tuning in, and we look forward to talking to you again soon. Thanks very much.
Voiceover: The information and views in this podcast are those of each speaker and were accurate when recorded. They should not be seen as financial advice.
[00:32:45] [END OF AUDIO]