Social Security Committee

Meeting date: Thursday, November 28, 2019

Official Report 557KB pdf

---

Citizens Basic Income Pilots

Item 4 is an evidence session on assessing the feasibility of citizens basic income pilots in Scotland.

The committee will hear from members of the citizens basic income feasibility study steering group, following the recent publication of its interim report. I welcome Gerry McCartney, a consultant in public health medicine at NHS Health Scotland, and Andy White, tackling poverty manager at Glasgow City Council. We have apologies from their colleagues Wendy Hearty, basic income project manager at Improvement Service, and Julie McLachlan, senior manager for economic policy at North Ayrshire Council. They have done a power of work on the issue, but cannot be with us, unfortunately. They have our best wishes.

Before we move to questions, perhaps one of our witnesses would like to make some brief opening observations.

We are grateful to the committee for the opportunity to speak to you. We welcome your interest in the steering group.

As members probably know, we have been asked to look at the feasibility of citizens basic income pilots in Scotland. That work has been on-going for some time now. The group was first formed in November 2017 and we were grateful to receive some money from the Scottish Government, as well as resources in kind from the collaborating partners, to take forward the feasibility work.

We recently published our interim report, of which I hope members are aware. It is an interim report because we are actively seeking feedback from a variety of stakeholders, including the committee. We will take any outstanding issues into account as we prepare the final report, which is due to be made available next spring.

I want to explore the rate that would be set, especially with regard to the minimum income standard. I am aware that you have proposed a CBI at two levels, and that the higher level is based on the minimum income standard that is produced by the Joseph Rowntree Foundation. Why did you choose that? What impact might it have on how much people would get compared with the current system? What might that do to tackle poverty?

The purpose of a citizens basic income that most of the collaborating partners—the local authorities and the Scottish Government—have considered is the reduction of poverty, but that needs to be balanced against the feasibility concerns with the different rates, levels and eligibility criteria that might be introduced.

I am sure that members will be aware of the different views that think tanks and organisations have about how a CBI should be implemented, what level it should be set at, who should be eligible and how the payments should be made. We had to narrow down the vast array of options and seek feedback on a couple of options that we thought were more likely to be able to be progressed.

We centred on two options. The first is a low rate, which would be more akin to the current levels of benefits. With that rate, we are interested in the additional impact that you might get from removing conditionality in the benefits system, which is one of the main proposed benefits of a CBI. People would not have to prove that they were searching for work and they would not have to undertake additional activities to be eligible.

The second rate is akin to the minimum income standard. As Alison Johnstone said, that level of income has been defined by the Joseph Rowntree Foundation, which looked at what it costs to live and participate fully in society. The minimum income standard is not just about someone being able to feed their family and clothe themselves; it is about their ability to participate in social activities and be a full member of society. It is quite a common definition of poverty. In pegging the higher rate to the minimum income standard, we would hope to vastly reduce poverty if a CBI were to be piloted or implemented.

The report says:

“The steering group has been engaging with DWP since April 2018.”

Will you provide insight into that engagement? How much interaction has there been between the steering group and the UK Government?

It is fair to say that our interaction with the DWP is not currently at the stage that we would like it to be. As you have said, our first contact with it was back in April 2018. The Scottish ministers were also in contact with it, as a result of which the Scottish Government wrote to the Secretary of State for Work and Pensions. That led to a commitment that DWP officials would be made available to interact with the work of the steering group. We had one face-to-face meeting with officials in March 2019, at which we set out our interim model and asked a range of questions of the DWP and its HMRC colleagues.

We have been planning a further face-to-face workshop at which we would go through the interim model and our questions on it, and I am sure that the DWP will also have issues to discuss. That was due to take place in December, but we have been told that it will have to be moved to January next year because of the forthcoming general election, which might be understandable. Representatives of the steering group will go down to London and engage with DWP representatives face to face at that workshop. We hope to have contact with them beforehand to discuss their interim thoughts on our model—or at least on the questions on it that we previously asked them.

On Alison Johnstone’s previous question about tackling poverty, I repeat what Gerry McCartney said. We have proposed two rates for the CBI, one of which takes account of the basic, simple subsistence income levels that people would get within the existing legacy system and the universal credit system. We have also looked at the minimum income standard that the Joseph Rowntree Foundation has set out.

The committee might conclude that the higher level would have a greater impact in tackling poverty—and some of our other work on economic modelling of the options might suggest that—but we should not underestimate the possible impact of the lower level. One of the factors that are said to have influenced the increase in child poverty is the current two-child policy in relation to both tax credits and universal credit. Even at the lower level, our model for a CBI would not have such a cap on payment.

You and Gerry McCartney have made the point about the CBI’s not being conditional. That would take away much of the stress for people and would also mean that more of them would get it, so even if it were to be set at the lower level it could have a greater impact than the current benefits.

I would like to probe that point a little further. If you are setting up a pilot that might substitute existing benefits then, regardless of anything else that you might need to do, you will certainly need to work with HMRC and the DWP and to have access to data and so on. It would therefore seem that your relationship with the DWP and your ability to meet it and to progress matters will be absolutely key.

Yes. As someone on the steering group who is doing quite a bit of work on benefits interaction, I am keen that we have such meetings with the DWP. I hope that the committee can see from our interim report that we are very conscious that we want to look at the CBI in relation to the other benefits with which it would have to continue to interact in reality. For example, even the pilot’s advocates say that it would still have to interact with the disability benefits system—although, to an extent, that system is changing in Scotland—as well as with the rules on housing costs, childcare costs and limited capability for work-related activity. We are keen to propose a pilot model that would interact well with the remaining systems, because we would not want to see one that would be to people’s detriment.

10:30  

That was an interesting line of questioning, particularly in relation to existing UK benefits. I get the point about the overlapping between UK benefits, devolved Scottish benefits and benefits operating under agency agreements at a UK level. I note from the Scottish Parliament information centre briefing that there is a recommendation for the elements of

“Current payments for disability, work capability, childcare and housing”

not to be impacted by a citizens basic income, although elements of benefits, such as

“income based JSA, income related ESA, pension credit guarantee”—

and child benefit would be removed, but I will come back to that in more detail in a second.

Clearly, it is critical that there is a discussion with the UK Government about how that works. I am conscious that the fiscal framework is about to be renegotiated. Our committee has focused on unintended consequences when an investment in Scotland for a citizens basic income does not have a clawback effect, whether on an individual in receipt of benefit or on the Scottish finances. I am sure that we would not want to have a situation where Scotland uses its money to invest in a citizens basic income if that displaces UK-reserved benefits. Have you done a bit more work on that, or are you waiting until you meet the DWP? That is the big thing that jumps out in all this.

The engagement with the DWP and Her Majesty’s Revenue and Customs will be crucial if we are to have more meaningful stuff to say on that. We are keen to say that the model that we are proposing is an interim model. Even within the steering group, as far as the low level was concerned, we wanted at least to replicate the basic subsistence levels. The steering group might have further thoughts on the proposed model. You have mentioned pensions, and that might be one example to consider.

The starting point has to be what the DWP’s position is, and not just in relation to the levels that we suggest should be replaced by a CBI. If the CBI has to interact with the legacy or universal credit system, how would that payment be viewed in relation to people’s income? We do not want people to be in detriment. Equally, we want to be able to interact with the remaining systems in such a way that people still get what they need for their rent, disability costs and so on.

The answer to your question is that I do not think that we are at the stage of giving the level of detail that you are asking for just now. It is important that we have the DWP response.

Is there an underlying principle that the individuals taking part in a citizens basic income would not be at detriment? Would there be an underlying principle that the Scottish exchequer should also not be at detriment in relation to displaced benefits? That is why I mentioned the fiscal framework.

It is important to recognise that what we are doing is a feasibility study. How feasible it will be to do a pilot depends on engagement with DWP and HMRC and on the cost implications for the Scottish Government. All those factors interact: as soon as you move one part of the jigsaw, other things change around. As Andy White has said, the report is very much an interim report, and negotiations and discussions are still on-going. Having this kind of conversation is very helpful for focusing our minds on the kinds of things that we need to resolve as we move towards a final report.

That is helpful. I am using these numbers for illustrative purposes only but, if the measure was to displace £100 million of UK-reserved benefits that would otherwise be spent by the UK Exchequer but is not spent, because a citizens basic income displaces that, we would want to ensure that the Scottish exchequer was recompensed for that. I was just raising that as a general issue.

I mentioned pension credit in my earlier question. Apologies if I have picked this point up wrongly, but my understanding is that the pension credit would be impacted by a citizens basic income under the potential model, but pensions—which are directly contributions based—would not be. I understand the logic for the latter part of that. However, although pensioners who are reliant on that top-up to their income, effectively for a minimum income guarantee with pension credit, might not be in detriment, they might not benefit that much, whereas a pensioner with a full contributions-based entitlement to their pension, depending on what rate you go for, then gets £167.25 or £195.90, and they would do very well out of it—and that is a positive thing.

However, could the unintended consequence be that that arrangement creates inequality between various groups? Could inequalities between different groups of pensioners be exacerbated under the proposed model?

Perhaps Andy White can go into more detail on some of the interaction issues, but it is also worth noting that, in a pilot scenario, we do not think that it is likely that income tax rates would be able to be varied. Therefore, there is a difference between what might be possible in a roll-out scenario and a pilot. The interaction between any changes in tax rates to fund CBI, were it to be fiscally neutral, could claw back some of the gains that more affluent groups might get from a CBI.

We have used a large chunk of the money that we received to commission econometric modelling that looks at different models for funding a CBI and what the redistributed consequences and knock-on consequences for the wider economy would be. You are right to say that the principal of universality means that everybody gets CBI; rich or poor would benefit in the first round. However, it depends on how a CBI is then funded, who bears the tax burden and whether other options might also be used to fund it. Those elements have as much impact on the distributive qualities of CBI as where the payments go.

In relation to your point about potential detriment and winners or losers, convener, as I have said, the low-level CBI payment model that we have put together looks at what could replace what you have defined as the basic subsistence rates. Clearly pension credit would be that rate for low-end income pensioners, and that is in tune with the rest of the model for the other age groups. As we say to an extent in the report, there is a good bit more work to do on how pensions are treated, because in some cases, if we look at the new state pension and the level at which it is set, we will see some pensioners getting almost double the payment of others.

We are going to have a further look at the issue of pensioners and the CBI, and it could be that we switch off the new state pension instead of the pension credit benefit. We might look at that—as many people do—as a basic income that is, to an extent, unconditional for those of pension age. If that is the starting point, we could also look at the state pension being retained and perhaps topped up. Those are among the options, some of which are in my own head. The steering group will take the final position on them. However, you will see from the wording of the report that in relation to pensions and how we think the basic element in working tax credits is treated, the position on pensions is an interim one, and we are going to do a bit more work on it.

I am going to ask you about the international experience, but before I do that, will you say when the econometric modelling that you talked about is going to be produced?

We have commissioned a group that includes the Institute for Public Policy Research and the Fraser of Allander institute at the University of Strathclyde. We have seen some interim results, and we had hoped to get a final report by January to include in our final report. We keep asking them to do slightly new things and to tweak things, so we might have to extend that timeline somewhat, but we are still to negotiate that with the wider steering group. We only met them yesterday. We are checking the assumptions and asking the research group to look at different sensitivity analyses, not least those concerning the issues that Andy White mentioned.

Will you be able to share that modelling with the committee once it is in a place that you are happy with?

Absolutely. It will be in the public domain once it is finished.

That will be helpful.

Tax and benefit systems differ, so international comparisons must be tricky. Systems elsewhere are moving from different points to other different points compared with our system, and they do not necessarily have to negotiate between two Governments. Nevertheless, it would be helpful to know how much international experience you have been able to draw on, particularly on the taxation system. I presume that citizens basic incomes are funded partly through a clawing back through taxation, so those who are more affluent who receive the income will pay an element back in taxation. Where has been the most fruitful place that you have looked at?

It is a really good question. A lot of work has been done in that arena. The steering group’s approach has been twofold. We have participated in the big international conference of the Basic Income Earth Network, which brings together people with an interest in basic income interventions, including academics. We have been interacting with the networks around that and we have had a lot of correspondence with others who want to learn from us, and vice versa. Through that, we have had a lot of contact with, for example, the people who ran the pilots in Finland and Ontario. I am using the royal “we”.

Wendy Hearty, who was unable to be with us today, is one of the authors of a large scoping review that was done with What Works Scotland and the University of Glasgow, which synthesises the literature from previous basic income experiments across the world and across time. I am sure that the committee is aware that there is a long history of such experiments, which goes back to work in Gary, Indiana and various other towns across the USA and the Alaskan experiment. There are also more recent experiments that are still to generate evidence, such in the Netherlands. I can run through some of the key learning that we have drawn from those experiments.

It is important to be clear about the aim of a basic income pilot. In the Finnish experiment, the aim was about helping people back into work, so the only eligible group comprised people who were on unemployment benefit. The basic income was not universal in the way that we are discussing in the Scottish work. It is important to be clear that different pilots and interventions across the world have different aims and scopes.

We are also keen to learn about the need to get a broad level of support across society for pilots. In a number of experiments, the learning has been lost because the experiments finished early. For example, in Ontario, there was a change of government and the experiment did not get finished, so a lot of the learning and research was lost. It is important that there is broad agreement about the undertaking of pilots and that they are seen through to the end so that the research can be completed.

There is a lot of learning on design. Some pilots have been undertaken at a local level, by municipalities rather than national governments, which has limited the scope to switch benefits on and off and change conditionality. Again, if that is one of the key aims of a pilot, it needs to have political implementation at the right level.

There has been a lot of learning on the impacts on particular groups. For example, there has been a lot of interest in whether people are more or less likely to withdraw from the labour market if they receive a basic income. We see from the evidence that, where that happens, the effect is small and it often involves those who want to stay on in education for longer and improve their skills. That has led to some interest about the potential for increased productivity in the future. Again, however, the extent to which we can generalise on that from pilots that took place 30 or 40 years ago might be limited, particularly if they were in a North American context.

The final thing that we need to learn is the importance of clear communications. In Finland, the global media descended on individuals and small communities where piloting was taking place, which created a lot of stress and pressure for the people who were taking part, and it might also have interfered with the true effects of the pilots. We would need to take care with all those factors, including media attention, if we were to go ahead with piloting in Scotland.

10:45  

It would be interesting to hear your reflections on what we can learn from the four local authorities. Are the differences between them just due to their locations? What is the communication like between the four studies? Are you ensuring that they talk to one another? Are they sharing information?

Just to be clear, I note that there are not four studies. We are all working together across the four local authorities. The Scottish Government is represented, as are the Improvement Service, which has a pan-Scotland local authority role, and NHS Health Scotland, which provides a lot of the evaluation support. No studies are in play at the moment; all that we are doing is looking in collaboration at the feasibility of a study. As such, all of the work is joint.

However, the four local authorities bring the different local challenges and demographics of their areas. One of the questions that came up in the interim report is whether there would be interest in piloting in an urban area, a rural area, a deprived area and a less deprived area. It would be possible to find a compromise area that has moderate levels of those different factors, but to get true community-level effects—we are interested in the potential for volunteering, social capital and other factors that might arise from the implementation of a CBI—we want natural communities to be parts of the pilot.

That is one reason why we would not want to pilot at individual level, with a scattering of people across Scotland. We want to have a critical mass of people who live alongside one another in order to see whether those community-level effects arise. That might happen differently in urban and rural areas and in affluent and deprived areas, so we are keen to get feedback on whether there is interest in that approach. Running pilots in more than one area would increase the cost, so there is a trade-off between the potential cost and what we might learn from the differential impacts in different areas.

I think that I am right in saying that you have suggested a three-year study, with a year to get everything organised. I presume that it would then take another year or two to analyse the work so, even if we pressed the button on 1 January, we would be looking at six or seven years before a final recommendation came before the Scottish Parliament. Is that the timescale? I am trying to get my head around what it would be.

The first thing to say is that we are some way off that stage.

We would not press the button until the necessary work had been done.

Yes—all the feasibility work needs to be done. You are right in that we have suggested that, from the point of starting, we would need a year to get things ready.

One of the main purposes of piloting is to learn from the impact, because there are so many uncertainties. To do it well, we need to ensure that the pilot is evaluable and that we can gather as much information as possible. That means collecting data from comparison populations as well as intervention populations, so that any secular trends—trends that happen over time in the general population—can be adjusted for. Collecting the data, designing surveys and implementing pilots all takes time, so we would want to have a year of preparatory time.

It would be possible to run pilots for shorter or longer periods. We suggest a period of three years because anything shorter might be too short for people to take advantage of a CBI. For example, people might decide to use it to retrain or start a business, but they would be less likely to make that change if they were only guaranteed the CBI for a shorter period such as six months or a year. If we were to implement it for 10 or 20 years, we would have to wait for too long to find out the impacts. Three years is a compromise that is based on the available evidence, although it is not fixed.

You are also right to say that we would need some time to analyse the data afterwards. A year or two would be a reasonable estimate of that. If it is an important issue, we can try to clarify it in the final report.

Realistically, when do you think that you will be ready? Whichever area you decide to go for, how long will it take to get the local authority personnel in place and to arrange everything else? If the Scottish Government came to you and asked when you could push the button, what would be a realistic timescale?

We are at the stage of exploring the feasibility of a CBI. We have already discussed some of the aspects that make it less feasible. We need to be clear about the benefit and tax interactions, and all of that would need to be resolved before we could recommend to ministers and others that we were ready to pilot. I think that, if all of that was resolved, a year from that decision point would be long enough to get things in place, but there is a lot of work to be done before we get to that point.

The pilot would take an area-based approach. As the politicians who are knocking on doors in our constituencies for the general election will know, some areas have a high churn or turnover with households moving in and out. Under a pilot, what would happen if a family moved to another area or a family moved in? Would the cohort that were there when you pressed the button, as Jeremy Balfour put it, keep the citizens basic income or would there be additions as folk moved into the area? Do you have any thoughts on that?

The proposal is that everybody who was resident on day 1 would get it and they would retain it no matter where they ended up, and anybody who arrived in the area would get it. That would allow some natural experimentation. How important is it to be part of the community? Does it matter if someone moves away? If somebody started on the CBI and they had to move for whatever reason, it would be unfair for them to have it withdrawn. It would also be unfair to incomers to the pilot area if they did not get it, and if we were trying to generate community-level effects, they might become diluted over time.

Other options could be pursued and the churn in communities could influence decisions, but that is our interim position. Again, we would welcome any feedback on that issue.

That is interesting. In my constituency, I can identify areas with lots of void properties or hard-to-let properties and clusters of temporary accommodation, so it would be fascinating to see the behavioural impact.

I want to explore the consent of participants. The report leaves that open, to an extent, although it talks about whether consent would be real or not. If existing benefits were withdrawn, participants with no means would, in effect, be forced on to the CBI. Have there been further discussions on whether a pilot should be mandatory or voluntary?

There are two aspects to consent: the first is consent to participate in the pilot, and then there is consent to participate in the evaluation. The evaluation question is easier, so I will deal with that first. As part of an evaluation, we would collect data from people in the same way that the Scottish household survey, the Scottish health survey and other surveys collect data, and people would need to actively consent to our collecting that new data from them. People would need to give informed, active consent to participate in the evaluation. That would be true of people in both the intervention areas and any comparison populations.

You are right to raise consent to participate in the pilot. We have discussed that at length in the steering group, and it raises lots of issues. As you noted, changes are frequently made in the social security system and the recipients of benefits are not asked for their consent, but the proposition that we are discussing is different given that it would involve a pilot and be set in a social research context. It would depend on the basis on which the CBI was being introduced and piloted. If it was more on the research side, consent would be more important. If it was more a policy pilot, consent might be less necessary.

If we were able to guarantee no detriment, we would be much more confident about not needing to pursue consent. Also, a pilot scenario is simpler because we would not be proposing changes to the tax rates, which people might be more inclined to opt out of if they had the opportunity. All those considerations are still live and real and we do not have a final position. You are right to raise the issue, and I hope that I have articulated our current thinking.

You touched on one of the key points, which is whether a pilot could offer a guarantee of no detriment. We know how difficult it would be to maintain no detriment, not only during the pilot but afterwards. What is the steering group’s view on whether it would be right to introduce a mandatory pilot where it could not be guaranteed that there would be no detriment?

I am glad that you have raised the issue of getting people on and off the pilot safely and in a position of no detriment. At the end of a pilot, if it was not to be continued or rolled out further, participants would go back on to the legacy system. Moving between the legacy system and universal credit has all sorts of implications for people, about which we need to be very careful.

Much depends on the negotiations and the agreements that we can reach with the DWP and HMRC. The report that we commissioned from the Child Poverty Action Group outlined in detail the potential areas that we need to look at in order to avoid there being any detriment. CPAG was keen for us to look at those areas in more detail, and colleagues have been doing that in order to minimise the risks.

We want to be able to say that participants will be in a position of no detriment, but we might need to put in additional provisions to make that guarantee absolute. The complexity of some household structures makes it difficult to guarantee absolutely that there would be no detriment using a simple arithmetic calculation.

We would need to provide a degree of protection for people who have migrated from legacy benefits on to universal credit, whether in relation to their disability premiums or to transitional protection in general. People’s entitlement under the legacy benefits is higher than it will be under UC. We need to tease out those issues with the DWP so that we get guarantees that people’s transitional protection would not disappear if they ended up coming off universal credit for a period because they were participating in the pilot scheme.

We have marked out loads of potential detriment, not because we want to be negative, but because we want to carry out due diligence. We have a sub-group to the steering group that looks at transitions and the support that people might need at the start of the pilot, during it and when they come out of it. We are keen to ensure that any potential support does not act as an intervention on its own that might shape behaviour, because it would then be difficult to work out whether the effect was caused by the CBI regime or by the additional support or intervention.

As Gerry McCartney said, Mark Griffin asked a good question. We are doing a piece of work on the issue and we are very conscious that we need to get it right.

Do you intend to seek agreement from the DWP, HMRC and Social Security Scotland that they would support anyone who came out of a pilot in order to ensure that there was no detriment afterwards?

We have set out a range of questions to the DWP, but you are right that there could be additional questions for the Scottish Government. The focus just now is on the DWP and on ensuring exactly what you have highlighted. How we go forward with a potential final model in a report will be shaped by the answers that we get to those questions.

11:00  

Will you offer your assessment of the model and how close it is to what we might call full CBI?

I will try. Most of the advocates of citizens basic incomes want them to be regular, individually paid, universal, unconditional and basic. The models that we are proposing meet those criteria and so would be fairly close to that full model. The bit that a pilot would not be able to look at is the changes to the tax system that would be required to fund it, particularly at the higher level, which would be more expensive. In that sense, a pilot would be less true to what it would be possible to roll out: there is a balance to be struck.

The model is the best that we can do, given that it must continue to interact with existing systems. Clearly, the characteristic of being unconditional is difficult when we say that in the model a CBI would not replace benefits for disability or the limited capability for work or work-related activity. Clearly, there is even conditionality around entitlement to housing costs. That is where the difficulty arises. There are two things to consider. First, what will the DWP allow to be switched off? Secondly, how can we guarantee that there is no conditionality when we are saying that it has to interact with other benefits?

You mentioned that the tax system will be essentially unchanged. Does that compromise in any way what you can learn from the model?

Yes. A number of things will be easier to learn from a pilot scenario and some things will be more difficult. The income distribution consequences will be more difficult to learn about, but that is why we are doing the econometric modelling that we mentioned earlier. That will give us a better picture of what the gross economy impacts of a CBI would be, were it to be rolled out. Used alongside that, the evidence gathered through a pilot would be really useful.

We are interested in a large range of outcomes: health outcomes, social outcomes, labour market outcomes and people’s propensity to volunteer or to set up new businesses. A whole range of things are of interest to various stakeholders and we hope to learn all about those. We also hope to do a number of qualitative studies alongside this, to get better insights into unintended consequences or unexpected things. We might even be able to design studies as we go along, to better capture those kinds of things.

Something else that people look for from a CBI is simplification. Given the limitations that you have just described, we will probably not get the simplification that they got in Finland, let us say. Nonetheless, are there things that you can point to or that policymakers can learn about how the system might be simplified in future?

Obviously, some aspects would be simplified. The benefits that would be turned off for a CBI would pool in the CBI, so there would be some simplification. You are right, however, that in order to avoid any detriment, there is a trade-off to made regarding retaining other benefits, such as those around disabilities, housing costs and what have you. We have been very mindful of the no-detriment principle and that has probably trumped some of the decisions that we could have made about greater simplification, because some people would have come out much worse off as a result of that.

Finally, I will pick up on Jeremy Balfour’s point. Are you satisfied that the four areas that have been chosen will give you a varied enough picture of the country? As I come from the Western Isles, I am afraid that my standards for rurality and remoteness are extremely high, and I do not consider the part where I live to be either rural or remote. I notice that some remote areas—for example, the Isle of Arran—would be covered in the four that you have talked about, but is there enough variety in the four areas that have been chosen?

To be clear, we have not chosen any areas for pilot sites; we are working with local authorities—

I meant the four local authorities—

No, it is okay—but there is a risk of confusion here. We are working with four local authorities to look at the feasibility of pilots. The models and the means of selecting pilot areas should be applicable to the whole of Scotland. It would be for ministers and others to decide whether pilots were feasible and, if they did want to progress to that stage, whether they were content to search for areas based on criteria such as how valuable and comparable they might be and the learning that could be gained from them. We would then select pilot areas accordingly. At the moment, I would not assume that any pilot would happen within the four local authority areas that have been mentioned.

Okay. Thank you.

I will continue with that theme. You are charged with looking at the feasibility of a pilot scheme. I am concerned about the utility of any such scheme and what it might actually produce. You are having to wrestle with difficult issues about consent and what might be possible in the context of the tax and benefits systems. That sounds a bit like having to clear the undergrowth so that you can have a proper seeding of something new to see whether it works.

It seems to me that you will have to look at the utility of whatever you propose and ask whether it would actually tell us anything at the end of it. Would both Governments not have to go all in on such a scheme, because that would give you the clearest chance of finding out whether it would work in the future—or is that just my perception? If either Government did not provide that support—perhaps because of the tax complications that have been discussed—and you did not get the simplification that Alasdair Allan mentioned, is it not likely that you would end up with a dead-end compromise of a pilot that would not really tell you anything at the end of it?

That is exactly why we have been asked to look at the feasibility of such a scheme. Our proposal for the best form of pilot that we could do would list not only what we might learn from it but what might confuse its effects, muddy the waters or make it difficult to learn from. If we were to lay those things out, that would help us to recommend the degree to which such a pilot would be justified.

Clearly, piloting would be expensive, but that cost would have to be traded off against what we might learn and the possibilities that it might open up. The more that we could learn and the clearer that that learning would be, the more likely it is that we would recommend piloting. If a pilot would not be feasible, for whatever reason—and especially if the learning that would come from it would be so unclear that it would not generate the utility that you have mentioned—we would recommend that it was not currently worth while.

On the point about the location and demographics of a pilot, I add that, as well as representing part of Stirling, I represent Clackmannanshire, which is the smallest local authority area in Scotland. It has higher deprivation levels per capita than most others, but also pockets of affluence. However, even there, the issue that the convener mentioned—about families now being quite mobile and being located between two areas, such as Clackmannanshire and Stirling in my case—would presumably produce a limitation on the pilot’s efficacy, as would the three-year period that has been mentioned.

Unlike Jeremy Balfour, I think that the issue is about long-term change in society and how we cope with it in future, so the idea of taking time to consider it does not worry me so much. However, even over three years, is it not possible that people might change their behaviour and move around to take advantage of the system in one way or another? For example, if they see that there is a detriment in one area, they might opt out of it, or if there is a benefit in another they might opt in there. Is three years the optimum time for such a pilot?

Again, there is a trade-off. From international studies we have learned that if pilots do not produce useful results after a reasonable period of time, they tend to end. The learning points from them might be lost because the final surveys or data collection—which illustrate the utility that you are interested in—have not been done. If our pilot were to be elongated, there would always be a risk that we might lose some of the learning that we might take from it. That said, there might be other options—for example, piloting over an initial three-year period but with an option to extend it. At that point we could collect the data and say, “Okay. This is what’s happened thus far. Would you like to continue?”

There are also options to use something called data linkage to follow people up over a longer period, although they would receive their basic income for only the three years. It is a bit more doubtful whether they would retain any changes in their outcomes once their benefits changed back to the previous system, and that confuses the effects.

You make real points, and I hope that we will reflect on them and will reflect them back in the final report.

Some of the potential benefits over the three years have not yet been mentioned. Productivity was mentioned, but there are also benefits for educational attainment, health outcomes, economic activity and levels of unemployment. Even with a period of three years, however, your ability to gauge the changes will be limited.

Yes.

Yes.

The interim report is really interesting. Keith Brown has just touched on the issues around the long-term sustainable change that you may or may not get through what is potentially a massive change to how we deliver support to our society.

In assessing feasibility, you are presumably considering whether the CBI would have a sustainable effect. In the report, you mentioned the pilots that are currently taking place across the world and the interest in those pilots, but my understanding is that most of them have ended now. Some have been ended early, for various reasons, and there do not seem to be many initiatives that have managed to sustain a pilot through, so as to deliver the kind of results that you have just discussed, noting what you have said about the interrupted data collection that does not provide sufficient depth. Could you talk a little bit about that, as the interim report seems to have cut before the pilots have?

Secondly, for any Government, the ultimate sustainability will be based on financial deliverability, even if a pilot looks good. I am looking at the numbers. You spoke earlier about how a CBI would be affordable in relation to the replacement of benefits and the recovery rate through income tax. Has any thought been given to where that would sit? You have talked about the basic income levels but, if someone is on £40,000 and gets the universal income of £20,000, are you looking to recover the whole of the £20,000 through income tax, in which case there would be a 50 per cent rate—using rounded figures—to claw that money back? Has any thought been given to that? Ultimately, as part of the feasibility assessment, the pilots must give some weight to whether the proposals would be realistically deliverable.

I will take your last point first. The financial deliverability of a pilot should be distinguished from that of a potential roll-out scenario. They are related, but they are different. A pilot scenario could clearly be costly. It depends on how many areas are involved, on the levels and so on. The negotiations with the DWP and HMRC are important, as some of the benefits are paid by the UK Government. If they are being replaced, would the money be recycled back to the Scottish Government to pay for the CBI, or would it be retained? Clearly, that changes the maths for how affordable a pilot scenario would be.

On the financial long-run deliverability, it is likely that tax changes would be required to fund the CBI. That is something that a pilot scenario could not consider. Even if we were able to vary tax rates for the individuals, that does not really tell us about the knock-on consequences for the economy. The best way to approach that is through econometric modelling—which has seemed to come up a lot this morning. We are putting quite a lot of faith in that, and we are examining a number of different scenarios for how the proposal could be funded and what the knock-on consequences would be for the distribution of income, for gross domestic product and for employment rates in future. All modelling comes with assumptions, and we can vary those assumptions, but that will all be transparent and will be shared with the public once it is ready.

Going back to your first point, about the sustainability of pilots, you are right that the evidence that we are drawing at the moment is from pilots that are finished, but there are quite a large number of pilots that are under way across the world. It remains to be seen whether they are sustained or not. Part of the rationale for pilots being stopped is simply that they had run their course. The Finnish pilot, for example, was planned for two years and ran for two years. It finished at two years, and the learning was gained. That is what was expected. There was a change of Government, and the new Administration did not support a roll-out scenario. That was fine.

11:15  

There have been other scenarios in which philanthropic organisations have run pilots. Again, that is a less sustainable model for roll-out; the purpose was more to generate learning and to create interest in the policy idea. At other times, pilots have been introduced at municipality level, but they have not been able to be rolled out further and sustained. North American pilots have often been stopped when there have been changes of Government. Sometimes the learning has been able to be captured, and sometimes it has not. A lot of the learning from the most recent pilot in Ontario was lost because it was stopped early.

I go back to my earlier point. One of the feasibility considerations is whether there is broad enough support—across political parties, civil society and others—to enable pilots to be seen through to the end, so that the learning can be gained. There not being such support would diminish the feasibility and utility of any pilots.

Have any of the pilots demonstrated that the policy is workable on the ground? If that has been demonstrated, why has the policy not been picked up by Governments?

By “workable”, do you mean whether the policy can be implemented?

I mean whether the policy could be delivered in the country in which it was piloted. Every country has slightly different circumstances. As you have said, a lot of pilots are being run and tested, but, as yet, nobody has run a pilot and then said, “Yes, this is great—we are going to go with this.” Why do you think that is?

There are some places where basic income-type policies exist. The famous example is Alaska, which uses a sovereign wealth fund model. That is done at a lower level, so it might not be that comparable to what we are talking about, but there are some examples of such a policy being maintained for a variety of reasons.

It is a shame that some of our colleagues are not able to be here today, because they are more familiar with some of the evidence than I am, but the pilots that I am aware of have all managed to be implemented safely and seen through to the end, unless there was a change of Government over that time. The question is why the pilots were not then rolled out. That is down to political choices and priorities, and the extent to which that policy approach is seen to be the right one, it having been balanced against other cost pressures.

One of the frustrations from a research angle is that some of the learning has been lost from the pilots. It would be interesting to see, when there have been positive early results from such pilots, whether those could be transferred to a Scottish context. All the pressures and challenges that we face in relation to poverty, inequality, health outcomes, productivity and so on have been mentioned, and they might be impacted by a CBI scheme, so it would be interesting to see whether there would be an effect in Scotland and whether we could collect data on that, if we could find a feasible model.

If the feasibility study shows that it is worth running a pilot, what are the main hurdles and barriers to delivering a successful pilot that would give us the data to enable us to decide whether such a policy would be feasible and sustainable in the long term for the Government?

The purpose of a pilot is not to show that something is a perfect policy and must be rolled out; the purpose is to learn what the impacts are. Everybody can then judge, on the basis of those impacts, whether it is worth continuing with the policy. Even if we were to find that the policy did not work, that would be useful learning, because we would know that it does not work. If we were to find that the policy did work, for whatever outcomes that people were interested in, I hope that we would have collected robust enough data and evidence. We will have a comparison group that will use high-quality study designs, and we will have planned for the pilot in advance.

If we manage to get all the feasibility ducks in a row, we will have enough time to do all that work, to get the research planned and to ensure that people get on and off the benefits safely, so that there is no detriment to them. That will give us the best chance of generating high-quality learning. Carrying out the feasibility study is an excellent opportunity for us to get all those ducks in a row, so that we are not left with too many uncertainties at the end of any pilot.

You see running the pilot being as much about feeding in evidence on how to deliver social security as it is about potentially creating a CBI.

Absolutely. We carry out pilots for a reason. If we knew that a policy was great and had no downsides or unintended consequences, we would not need to pilot it. The policy is an interesting idea with lots of potential, but we need to run a pilot so that we can evaluate it carefully and be sure of what the impacts will be.

We are almost out of time, but I have a brief question. Earlier, we were clear that four geographical areas will be selected for piloting a universal basic income. Was consideration given to picking certain groups within society to see how a pilot could operate for them? Before the meeting started, I mentioned kinship carers, people who are living in temporary accommodation and lone parents. We could pick a much wider area or a certain section of the community to see how introducing a citizens basic income could improve people’s life chances, health and outcomes. Was thought given to going down that road? Has that option now been ruled out altogether?

We thought about that option, and some local authorities were particularly keen on it. Care leavers was another group that was mentioned. There is a case to be made for running a pilot involving all those groups. However, by taking that approach, we would not gain any idea of the community-level impacts that there would be. Those are some of the most uncertain impacts from previous basic income approaches, and there is a high degree of interest in those outcomes from the stakeholders to whom we have spoken. On that basis, we cannot have the population dispersed in that way; we need to have a population concentrated in a specific geographical area.

Before we move into private session, does Mr White or Mr McCartney want to make any final comments?

We welcome the opportunity to come along today. As we have highlighted, this is an interim report. We are not saying that everything in it is absolutely correct. We have asked the stakeholder group and others to make comments, and the report has been made publicly available. We would be very interested in the committee’s comments on the report.

It has certainly been a thought-provoking and interesting evidence session. I thank both witnesses for their time and for giving evidence.

We move to agenda item 5, which we previously agreed to take in private.

11:22 Meeting continued in private until 11:37.