 Mae'n ymdgeisio twisting 메ite! Mae'n not arrivedeb y soldiers cylliad a'r borhagiau newid yn nodi'r y p 간단 sy'n ei ddag ei gy receiving plate m todavía yn y medd Twiceagaf. Cyflawnoddi, gyda'n gweld mewn i'r 느낌이erdem ac'r magazine nifer o'r myiser a'r yn dechidedd y Cymru ac yn dda i aml回來ol mwyaf an Smartphone, i bern om dinosaurs yn gweld i chiicado'r tre lever, oedd Aunt Grace yn handfulwch â ddegosламwysau We bell Professor of Economics at the University of Stirling, and Dr Jim McCormick, the Associate Director of the Joseph Rowntree Foundation. Welcome, gentlemen, and thank you for coming along this morning. Can I just kick off in the area of no detriment? I've been thinking about the potential of any future welfare changes from either the Scottish Government or the UK Government after the full package of the Smith proposals become law and wondering in reality how workable they are, particularly around the principle of no detriment and the principle 2, which states that there should be no detriment as a result of either Government's policy decision post-devolution. Is that really possible within the scope of the terms that it's meant, without areas for potential disagreement, and is it realistic? I just wanted to have a better feel for that, because I've had some evidence of it, but I've not really got underneath it though, so I don't know who would like to kick that off, David, Jim. The simple answer, I think, is that it's not possible to enact the second part, the second form of the no detriment principle, without going into massively complex calculations with the huge potential around that for dispute between the different authorities. For example, the air passenger duty and Newcastle airport impact has taken as a kind of classic example of how you end up tying yourselves knots in trying to figure out what the impact on the rest of the UK Government would be should Scotland decide that it wanted to reduce its air passenger duty or eliminate the air passenger duty, because you've got to try to imagine what would have happened had Scotland not taken that decision. It's around that question, which we call a counterfactual, that all of the uncertainty arises because it didn't happen, so how could you know what would have happened had Scotland not taken the policy decision that it had taken? You end up in a situation of, I guess, mainly economists arguing with each other about what the impact might have been had the decision gone a different way. It's not clear that there's any easy way to resolve that without long, well, maybe not at all, but certainly it would involve long and complex calculations. I looked at the House of Lords' work on that and they interviewed a man from the IMF, called Carlo Cotarelli, who said that there was no similar principle in operation anywhere else in the world, so that's the basis for my argument that I think it's unworkable. I think that the fact that we're in uncharted territory internationally, as David said, as we heard in the last session of the committee as well, is really significant. If we are to pursue this principle, I think that it's really important that there are some really clear principles that are underlying it about transparency and, as far as possible, simplicity. Transparency is about working methods, so it's not about one Government trying to pull the wool over the eyes of another Government. Proportionality cannot be the case that no detriment means that minor changes in budgets at the margins are given the same prominence as the very major incursion into, for example, tax territory, so income tax will be a shared tax base between the Governments. We have to understand that, in a shared tax base, there could be substantial consequences of one Government's actions versus another's actions. I don't think that we should give up on the principle. I think that we just need to have some proportionality and be able to distinguish between major and minor detriments, which may, in fact, net off each other, plus and minus either side of the board. I'm in favour of shining a light upon what appeared to be the major risks and perhaps benefits for each Government, building expertise to understand those major budget consequences, which, as I say, come from both tax decisions as well as welfare spending decisions, potentially, and build up our data and evaluation around those more major areas, rather than fretting around every potential consequence in our budgets. I agree with Jim on that point. If there was a major decision about the size of the state, which had a major impact on, let's say, Scotland's block grant, the rest of the UK decides that the health service should be privatised, which would have a major consequence for Scotland's block grant. At that point, I think that you want to have something kicking in to negotiate what are the implications of that for Scotland's funding and going through that. It does kick into what you probably will also want to discuss later on how the block grant is adjusted and the way that it might be adjusted could change in the light of those kinds of decisions. Malcolm MacDonald wants to ask questions about block grant adjustment before I get the other. I think that that plays in, David, something that you said in pages 11 and 12 of your written evidence, where you say that, if attendance allows in England was to be devolved to local authorities, there would be no relevant data on which to calculate Scotland's block grant adjustment for this benefit. Therefore, either of our witnesses today are aware of any proposal to change the structure in England of benefits that are proposed for the devolution that could have that kind of impact, because there's a sort of area, I think, that would be a big change. This is purely hearsay that I've heard when I've been down south that that's all, but it is true that local government's social care budgets in England are extremely cash trapped. They might be making a case to transfer some of the disability benefits budgets into the local government budgets to build up their social care support. Then, I guess, it becomes a little bit like council tax benefit, which I mentioned in my paper as well. That has been devolved down to the local authority level in England. What has happened there is that the local authorities can no longer identify that part of their grant from central government, which is associated with the payment that they got for council tax benefit. If the same thing happened around attendance allowance, there really would be very little to compare Scotland's attendance allowance. So how could we go about—if that was to happen—if there's no data being able to adjust the block grant appropriately? Would there be a way? The only thing that occurs to me is that you could—another thing that I mentioned in the paper is that you might index to the so-called at-risk population. You have estimates of how many pretty severely disabled people there are, because attendance allowance covers that kind of condition. In Scotland, as against the rest of the UK, how are those two moving relative to one another? That would be a relatively objective way to assess the potential for attendance allowance claims. Of course, claims behaviours differ. It's not just disability that matters who claims and how they claim. Whether they're encouraged to claim also makes a difference, as far as the ultimate cost to your budget is concerned. The bottom line is that if DWP decides to do this, it's going to make it more complicated in terms of sorting out Scotland's block grant. That's going to be inevitably the situation. It's something we need to be aware of when we draw a report up, so I'm glad we don't draw it. Do you want to make any comment on that? I think that we can perhaps anticipate over the next five years an interest in Treasury and DWP to move further in the direction of localisation of some benefits in England. How far that goes is, of course, unknown, but you can detect a broad interest in doing that. A complicating factor is variable localisation with some cities in England having a different deal from the rest of the local government in England, and perhaps even the city deal in Scotland having a different relationship with the Treasury. The picture that emerges is one of genuinely variable geometry or variable deals for different tiers of government. The only way through that, I agree with David, is to look at aggregation factors affecting potentially eligible populations, whether by age or based on historical data. It certainly means that we need to be using three to five-year rolling averages rather than caching out in a single year. We'll go and see that bit again, will you, please? Measuring populations that might be eligible and the budgets attached to them, but over running averages of three to five years, not taking it a single baseline year as the best indicator we've got. This is a bit indexationally in the future. It's really important that we have those broad rolling averages as our measurements, because they can help to incorporate changes in data from year to year, which are perhaps not typical. If we go down that route, and it seems a sensible route, you've got to have the data to provide that. I was discussing the issue of data yesterday in the presentation. In 2001, the Labour force survey, which collects some disability and health-related information, had a sample of 10,000 each quarter in Scotland. Now it's down to 6,000, just over 6,000 a quarter. There needs to be some kind of guarantee that the data, if you go down that route at looking at eligible populations, is adequate and that you don't spend your time worrying about the quality of the data when big decisions are being made. I know that Linda wants to ask questions later about individual detriment areas, but before we go there, it probably naturally flows into block grant adjustment at this stage from that discussion, because obviously we have to have some method if we're not going to go into individual negotiations of adjusting that block grant adjustment. Malcolm, I think that you had some questions. That was really interesting. I think that the theme of your presentation is the complexity of that, and it could become more complex because you're talking about attendance allowance, but that's been skewed in Scotland because of the free personal care issue that we all remember. In terms of the changing populations, it was interesting because I asked this question last week, and I don't know if you've seen the official report from last week, but David Phillips of the Institute of Physical Studies just cut to the chase in that kind of way. I assumed that the method that has achieved a certain amount of consensus for tax in terms of per capita index deduction would not be applicable to social security, and he kind of surprised us by saying, well, just use it, because if you try to do something else, there are all sorts of problems attached to it. David Bell mentioned this, and interestingly, you pointed out in your paper that per capita index deduction wouldn't work so well for us. It's the opposite of income tax, so index deduction would be better. However, I suppose that the question cutting through all that is, is it reasonable for us—I mean, it's understandable for us that we just want the best deal for both, but given the difficulties that we will have to get our desired objective for income tax, is it then reasonable that we work out some complex method that's going to give us the best deal for social security? Perhaps it's understandable and reasonable, but is it that risky or that bad just to go for the per capita index deduction for social security in the same way as for income tax? I think that there is a question about the deal that will be done and whether, in order to get a deal done, a simple deal is done. If a simple deal is done, what will happen is that one of those methods will be chosen and it will apply across the tax and welfare horizon. In those circumstances, as far as Scotland is concerned, it's more important to get a good deal on the tax side than it is on the welfare side, just because there's £11 billion in income tax alone, and what we're talking about here is around £2.5 billion. The costs are going to be greater overall for the Scottish budget if the income tax deal is less beneficial as far as Scotland is concerned. That would, of course, just take away all the issues that we've been discussing about indexing on relative populations at risk and so on. Partly it's a question of what risks are being shared across the UK as a whole and what risks are not being shared. Really, if you focus on the tax block grant adjustment, what you're saying is that the risk that you're looking at is Scotland performing economically at a different level to the rest of the UK. If you go for indexation and wealth, that's the risk that you don't cover with the block grant adjustments that are being proposed for income tax. If you go for indexing the welfare separately, what you're saying is that the UK is willing to cover changes in population or changes in disability in Scotland because the grant to Scotland would increase if Scotland became relatively more disabled. In other words, it would have an insurance policy against increases in disability. None of that was the issue of those kinds of principles, which was never ever set down as the negotiations went on through the various stages and indeed in the bill. We're kind of left scrabbling around now trying to figure out which things it would be better to share at the UK level in terms of what Scotland would get out of it and which would be better not to share. That's a very long-winded answer to your question. My basic point is that if a simple deal has to be done, then the focus has to be on the tax powers rather than the welfare powers. Are you coming in on that or not? I suppose that going back to your initial thoughts on relative populations at risk, I suppose that one of the arguments that's used against that is that it takes away the incentives in terms of the relationship between disability and health, but that may be marginal since a lot of people with disabilities, that's difficult, but I suppose that the other thing is the distribution across the population. Can we predict that for the future? You say, David Bell, that the main client group for the welfare benefits being transferred to Scotland is older people. Is that in fact the case because the biggest benefit by far being transferred is disability living allowance, which is people under 65, so is it in fact older people, unless you've got a rather broad definition of older people? The disability living allowance is awarded based on your age at the time of your claim, but if, subsequently, you pass the age of 65, you will continue to receive disability living allowance. No, disability living allowance. I thought it became attendance allowance at 60. No, it doesn't become attendance allowance. The attendance allowance is paid to those aged over 65 who become in need of personal care subsequent to reaching pension age. Disability living allowance can be paid to those aged—no, can only be paid initially to those aged under 65, but if they, in the course of their claim, go over the age of 65, they continue to receive disability living allowance. I think that one of my graphs shows that the majority of claimants of disability living allowance have been claiming for a very, very long time, so quite a big proportion of them are aged over 65. I have about one third of the budget by budget terms. Because it has been suggested before that you are suggesting it, you could index it to the elderly population, and people have this view that the elderly population in Scotland as a percentage of the population is increasing more in Scotland, although I think that there is some controversy about that, but anyway. You could argue if it is skewed as you are saying to older people that you could use the numbers of older people or would that—because it is going to be difficult to get the specific information that you want about the relative population at risk if you have said that yourself. Would that actually be your ideal method if you could actually have an accurate assessment of the relative population at risk? You would use that and just keep updating that every few years. Would that be your ideal scenario? That is the ideal insurance policy. It does not provide you with the incentive necessarily, as you mentioned, around introducing public health measures to reduce the level of disability in Scotland relative to the rest of the UK, but if you want to be covered to get adequate money from Westminster to cover the current population of disabled people in Scotland, then you should look for one of those kind of objectives. To what extent would the number of older people be proxy for that or would that give a quite different situation? I am not sure that it is all that good. I think that you might want to fine-tune it a bit. Scotland's share of attendance allowance payments has been falling at the same time as, apparently, Scotland's elderly population relative to the rest of the UK has been growing. There is something down underneath going on there. I show, for example, that the number of claims related to frailty has suddenly dropped off—the number of attendance allowance claims related to frailty. Is that partly because of older people in carerums or is that marginal, in fact? The population in care homes in Scotland has not changed at all since the free personal care legislation was introduced. It is not that, I do not think. They are disqualified from attendance allowance, so I presume that you would have to take account of that. Those will be people at home living at home. The picture is different for each of the benefit areas being devolved. For attendance allowance, the population measure is a very good measure. For DLA PIP, it is a very partial measure. We will look at over 65 for the reasons that we have been discussing. Incentives are important to keep hold of that principle and map out where we can build incentives to get the benefit of our investment in Scotland. It looks across different areas. We should add in, perhaps we will come on to it later, the importance of employment programmes and income tax revenues. In earlier adulthood, we are building in proper support for people to be in better health working, which flows through long-term into how people reach older age. Just on demography and Scotland's place, Wales has the oldest population structure in the UK. Northern Ireland has the youngest, along with London's. Scotland is at the older end of the spectrum, but many parts of the UK, leaving aside the extremes, are converging over the next 30 years. That is before we build in migration assumptions. I am less confident in the view that we know how Scotland's share of the GB or UK—older people's share—is going to look over time. We know how it looks now and in the past. We can do some short-term projections, but we have to be careful about some of our assumptions. Two people want to ask on supplementaries in the city of Mark and Stuart McMillan. Mark is not mentioning David Phillips of the Institute for Fiscal Studies. Last week, he suggested that, because welfare spending is an AME budget line, it would be easier to adjust for that than it is for the Dell budget lines. Listening to what you have been saying so far, it does not strike me that that is necessarily the case. What is your view on that? I am not quite clear what is going to happen. I looked at council tax benefit and what had happened in relation to that in England. That is one instance where I know that money has gone from the AME budget and into the Dell budget. It has gone into the Dell budget and then into local government individual budgets and then it is in competition with other parts of the local government budgets in England. If they have a council tax benefit programme that is very cheap, they can spend more on education. It is not clear to me as yet, although I was assuming that that would be the case, that that welfare budget would form part of a Dell allocation to Scotland rather than an AME allocation. In that case, you are then opening up this question of what extent would you wish to redirect monies to local authorities to support social care rather than to attendance allowance? David Phillips might have more information on that than me, but I think that AME to Dell conversions are not unknown. The implication of that is that those budgets are in contention with other parts of the Dell budget. It is a context point, which is that the welfare cap is still something to watch. We know that the UK Government has made a statement on that it cannot be accountable or responsible for spending in Scotland on those benefits, but the welfare cap accounts in this financial year for well over half of all welfare spending. Even if the UK Government breaches its own target, as OBR thinks it will in three of the next five years, it is still an important bit of the Chancellor's thinking. I think that eight or possibly nine of the benefits to be devolved to Scotland are currently sitting inside the UK welfare cap. I would add that as a contextual point to watch, and it should be part of the negotiations and fiscal framework. Even if it is easily resolved early on, it is something that may come back in future and we should be alive to where that debate is going. Obviously, there has been well documented evidence and coverage around take-up rate and the fact that a large amount of potential welfare spend goes unclaimed. I have called on the chamber already for there to be a look at how benefit applications are handled and whether there are ways that that could be simplified that might improve and increase uptake of benefit. Now, if that was to happen, how would you see that being factored in in terms of block grant adjustment? Obviously, that flows from a policy decision rather than a situation at the point of devolution, but it would potentially have a material impact in terms of welfare expenditure relative to the rest of the UK. The ideal position would be to have forecasting and out-turn measures that are getting as close as possible to underlying eligibility for benefits, which are building in incentives to raise take-up over time. Some take council tax reduction, which has already devolved. Our Scottish data and take-up are not good, but we can assume that we still have a problem. If it is equivalent to the GB problem, we are talking about one-third of households in Scotland who should be entitled are not claiming council tax reduction. That is a major undershoot on what the budget ought to be. You can do similar things across other benefits, both for working age and retirement. I am in favour of, over time, a fiscal framework that gets better at measuring what underlying eligibility should be and trying to close that gap in terms of take-up over time. What is the likelihood of that? At the moment, we have a very clear ideological position at a UK level, which is looking to reduce the amount that is spent on welfare. That would tend to suggest that there would be less of an incentive to drive uptake of benefits and to ensure that those who could take uptake up benefits did so, because that would increase spend. What is the likelihood of a UK Government agreeing to a position that says that we will account for increasing uptake because we can then measure the number of people who should be getting benefits and account for the increased expenditure accordingly? The likelihood is that more of the heavy lifting will be localised in England. I think that we will see the UK Government insofar as they talk about take-up. They will be talking mainly about budgets and capacity for local government to do most of that. The exception is universal credit. If we ever got to a position in which that was rolled out with the right real-time information and the right design principles, universal credit could substantially raise take-up because it is a basket of those benefits. Across those six-bit suspending, which are currently separate, that is where the potential to substantially raise take-up lies. In other areas, in council tax reduction Scotland already has the responsibility to raise take-up. I think that we are well below what the budget should be for that. Other benefits will be reserved and, therefore, the responsibility will lie with mainly the UK Government. Although we know that the evaluation evidence tells us that the most effective take-up campaigns tend to be face-to-face local through trusted intermediaries such as housing associations and well-known charities. That work does not come for free. The goal of what needs to be done is very clear, but achieving that through fiscal framework is long-term work that will never end, but we should be at least starting at the right place and putting it on the table as an important principle, even if it takes a very long time to make progress. Maybe David, you take a view on this. How often or frequently do you think that this needs to be subject to review in terms of the arrangements and how effectively they are working? If we are seriously looking at some indexation towards the welfare powers, then take-up is an important part of the overall picture that determines the budget at the end of the day. Patterns change the costs to individuals of applying for different benefits over time. Summer difficult, summer much less difficult, winter fuel allowance and the state pension are almost 100 per cent take-up, and you find other benefits that are more complex that people do not understand or are unwilling to go through the relevant tests in relation to them. The take-up rate drops quite substantially. I know that some of my colleagues down south have done work on what are the factors that influence take-up rates and maybe that is something that we should be aware of, but it seems to me that it is such an important part of the determination of the overall budget that it is something that you would have to revisit on a fairly regular basis. I do not see how you can forecast how it will change. We do not have the information in the first place about take-up rates, because we do not really measure eligibility all that well now, so how are we going to know where it will be in five years' time if we do not know where it is now? I do not know whether it will be helpful to us if we could apply your proportionate test to some of the issues that we are discussing here on a risk register. If we apply the proportionality test that you referred to earlier to this area and the level of risk here, because I find it difficult to get my head round some of the stuff about where are the really, really important areas and where is the big risk? If every issue is presented is very, very important, we will lose our focus. It is important for us, when we finally get to our report, to take up some of your caution about proportionality and risk here. How important is this area and the scheme of things as a balance of risk for the age and population, which is still retained by the UK in terms of pensions, for instance? Where is the balance here? We can say that the persistent problem of low take-up at working age and older people in Scotland across the UK and the world, but in Scotland it runs into hundreds of millions of pounds of unclaimed benefits to which people are probably entitled, so hundreds of millions of pounds. If you then compare that with on the tax side, let's take the example of the UK Government's plans to raise the personal allowance and other tax thresholds faster than inflation over the next five years. Those account for hundreds of millions of pounds being lost to Scottish revenue, so I would suggest that those are major in terms of the share of budgets. I will give you one more example, which is much smaller than some of money, but it is very important for people affected, which is another test of impact. If low-income households in Scotland, as elsewhere, are going to lose universal credit income, for example, by scrapping other work allowance for single people, that damages work incentives and reduces incomes. It has a shunting effect potentially into the council tax reduction as incomes go down, and its eligibility for an already devolved benefit goes up. That is not probably a huge budget impact, but for the people affected it could run into hundreds of pounds a year, so I think that we need to test the proportionality, which is about the global budget, and then some kind of understanding of what is the impact on households affected, even if those numbers are not huge by the share of the population. Can I tease that a wee bit more out before I go to Stuart Millan? You mentioned tax allowance issues, or tax-free personal allowances. If the UK Government plans to increase tax-free personal allowances to £12,500 and to raise the higher income tax threshold from £43,000 to £50,000 by 2021, what would be the cost to the Scottish Government's budget of not implementing that change? That becomes a real issue of policy difference. I can go back and run my model and give you an answer to that question. I do not have it now. What I cannot give you the answer to is how people will react. I gave evidence to the finance committee on the issue of behavioural responses to changes in taxes. If the personal allowance goes up considerably in the rest of the UK relative to Scotland or the higher rate allowance, how does that influence where people choose to work? We just do not know the answer to that question. It is very difficult to speculate how that might be. That is a second detriment effect. A cautious Government will end up tending to follow what the rest of the UK Government is doing. Before you go and do your modelling, I want to get a feel of it, but we are talking about hundreds of millions of pounds here in the way that Jim is describing. If that was to happen, it is substantial. GRF has supported IPPR to boost its tax-benefit modelling to a much bigger sample size for Scotland. We will be reporting broadly in this area within the next few weeks, so hopefully in time to be helpful to your inquiry. If you turn that round into a question of what could Scotland do with future powers to either mitigate or to take a different view from what future Governments in the UK might do, well, we could choose to raise the basic rate limit with inflation rather than above inflation. We could choose to do the same for the higher-rate threshold or even freeze it. Those sound like quite minor adjustments, but they generate hundreds of millions of pounds of revenue. I am not going to put a figure on it, but it is certainly in that territory. Choosing not to raise faster than inflation would be a significant decision. Similarly, if we chose to raise tax thresholds at a higher rate than the UK Government, on the contrary, it would cost Scotland hundreds of millions of pounds. For us to be very clear about the revenue consequences of tax allowance and threshold decisions in future, as well as for spending, are there two sides of that coin? You are indulgent, so I want to tease it a wee bit further, because it is an important area, if you do not mind. I know Malcolm has a tiny supplementary as well. We are talking about welfare there, but what about the ability on the other side to raise tax and what that can do in terms of where we are? Spice has undertaken some modelling in this area of income tax. If we could be raised in 2016, for instance, if one P were added to the higher-rate band of income tax, I mentioned that. I want to compare what might cost under some of the changes that the UK Government has made under its austerity policies. Could you make a stab at what might cost reverse some of the changes to universal credit, the disability benefits being devolved to Scotland, health, childcare, funding, proposals not to implement that proposed income tax threshold that we are talking about? Are the numbers just big again? If they are, what sort of timescales would those big numbers be over? I want to enter a huge cautionary note about estimates of changing the higher-rate of income tax. First of all, we do not know enough about trends at that rate, trends in taxpayers and how incomes are holding up higher up. We do not know enough about behaviour responses. I know that there is a base model for trying to estimate behaviour responses. I would be veering towards the lower range of those estimates. The midpoint of a lower range of the estimates is around £75 million to £80 million of additional revenue from £1 on the higher rate, £1 on the basic rate, where we can be a little bit less cautionary, is the midpoint of the ranges about £325 million to £130 million. I am talking about ranges and midpoints deliberately because I think to pick a figure is unwise. That is about revenue raising. Decisions on basic rate limit and other tax thresholds can either raise revenue or take up revenue, depending on our choices, but those are again big sums of money in the context of certainly compared to the £2.5 billion in the welfare budget that may be coming our way. On the other side, if we were to mitigate against some of the welfare changes, what sort of costs would we be talking about there? I can give one example. I have already mentioned abolition of the work allowance for single claimants within universal credit and for couples without children. The debate has moved quickly from tax credit reductions and mitigation of those because of the autumn statement. I think that understanding the impact of abolishing the work allowance is probably the biggest and most urgent issue that we now face in terms of universal credit. The risk is that, by removing the work allowance, we start to undermine the purpose of universal credit. Without getting into that, we should be very concerned about working centres in Scotland. However, fully reversing the work allowance cuts in universal credit depends on how quickly we roll it out. However, let's say that we were fully rolled out by 2020 and that all the eligible singles and couples in Scotland were subject to that cut. It is again hundreds of millions of pounds and it is within the region of £200 million, so revenues lost and therefore cost of mitigation. There is a whole other debate about the technical feasibility of mitigation, how you interact with the DWP, as it will be then, who will take over full responsibility of universal credit. Over and above the mitigation budget, I cost of using those systems and allowing variation in UK systems. However, that is a ballpark figure. I think that the IPPR will give us a much more accurate figure within a few weeks. It is hundreds of millions of pounds of work allowance, so we are talking three pence in a higher rate right away, effectively. That is a significant impact on taxation. I agree that all those estimates are subject to considerable uncertainty. You can turn the handle and figure out what the implications are if people did not change their behaviour at all, what the effect on tax revenues would be, what we are not good at and there is relatively little information on what behavioural change they might make. We have to be a bit concerned about how high earners respond, because they contribute such a big proportion of total income tax revenues. There is some work by Alan Manning at LSE on what the effect of the increase from the 45 to 50—I have forgotten when it happened—the change in the additional rate between 45 and 50. It is not very clear that an increase in rate added anything to total revenues. You have to be cautious, because you do not know what the behavioural response might be. Jim Smith says that that is less of an issue down the income scale, where people might have less opportunities to find ways of avoiding tax. I will need to move on to another area. I will come after that. We will come back to that later. There will be decisions also for Scotland on the basis of the point on which we choose to uprate devolved benefits over time. That is separate from the indexing of the budget coming into Scotland, but a choice over how we then deploy those resources over time. Do we want to continue with a freeze in working-age benefits? Do we want to link to inflation? Do we want to link to some measure of median earnings? Those are big choices that all have a very big band width in terms of the budget implications. One of the 100 million that you were talking about was the UK Government raising the thresholds beyond inflation. Are you confident that whatever version of block grant adjustment we get, that will be fully compensated for? If Scotland is indexed to the growth in the rest of the—the Scottish block grant adjustment is indexed to the increase in tax revenues in the rest of the UK, the tax revenues in the rest of the UK would increase by less because you are adding to the personal allowance so that people are paying less income tax. As a consequence of that, the block grant adjustment would be smaller, which means that the amount taken out of Scotland's block grant from Westminster would be less. Scotland would in a sense have less taken away from its budget due to the increase in the personal tax allowance south of the border. We should not get too alarmed about that particular aspect, although it sounds alarmed. It does sound alarmed, and I am not entirely clear about exactly how it would pan out. Again, there are behavioural implications, but there is probably a second order. In the first order, there would be some cover for the change. I am keen to establish your thoughts and opinions on the issue of the block grant adjustment and the node settlement 2 principle on the welfare powers that are to come to the Scottish Parliament. Benningham and Smith recommended that the powers that were to come here, the Scottish Parliament, to have the complete autonomy to determine the structure and value of the benefits or any new benefits or services that might replace them. If, at some point, a future Scottish Government decided that it wanted to maybe scrap one of those devolved benefits and introduce something else, how would you think that would play out in terms of the block grant adjustment, no settlement 2, but also, as Dr McCormick reminded us of earlier, the fact that those benefits are included within the UK Government's welfare camp? That is a complicated question. I guess that it will depend on the agreement. If the fiscal framework agrees a single mechanism for adjusting the block grant across taxes and across welfare, the decisions that the Scottish Government might make in relation to welfare, although it might affect the welfare cap, presumably it can just make, because the block grant adjustment would be indexed on income tax revenues in the rest of the UK. Once you start fine-tuning down into the welfare indexing in relation to welfare powers, you end up in the same problem that we started with if Scotland were to abandon one of its benefits. Is the mirror image to whether the UK Government decides to abandon attendance allowance, which is where we started? You end up with the problem of trying to compare what would have happened had you still got it. That is a difficult one. Can we just index to populations at risk taking account of whether take-up is different as between Scotland and the UK and what allowance you might make for that? Although it is a complicated question, it is a mirror image of the question of what the UK Government decides to do with its welfare powers. Ultimately, when we know the fiscal framework, which we will hopefully do in a week or two, we will be able to come to a much better answer to the kind of question that you are seeking and answer for. We should know better then, but if the structures change as between Scotland and the rest of the UK and you want to index on welfare powers, then you have to have some outside way of calculating what would have happened had the welfare powers not changed. It makes the argument for us clear and transparent an indexing formula is possible, which in principle gives Scotland maximum space to decide how to use those budgets without worrying about its consequence of scrapping or renaming or redefining a particular benefit. Some of those spend decisions go together in families or clusters. If you think about the way in which DLA or attendance allowance can act as a gateway to carers allowance, the fact that those three come together in Scotland is helpful because the consequences go in a loop, if you like, and they are internalised. There are other decisions that will spill over. If Scotland made fundamentally different choices on DLA PIP to the rest of Britain, that might have a consequence for disability premiums within universal credit, which is a UK budget line. There will always be internal and external spill-over effects. On the point about who scraps what, a live issue is in the autumn statement, the employment programmes that were in the spirit of the Smith commission, work programme, work choice, are all but being scrapped now. What we thought were going to be particular budget lines with budgets attached to them is an 83 per cent reduction projected around those budgets. That is a live issue already before we even get into the detail of fiscal framework as a result of the chancellor's decision. Now, you could then say that those programmes are being all but abolished and that the budgets are being fundamentally reduced. Jobcentre Plus, a UK wide, GB wide network, will take on some of those responsibilities. That should trigger a question. Can we therefore track the total resources that are being spent on the same client groups that were intended for those programmes? To what extent is a great area there where we are negotiating exactly what is being devolved? Is it in the spirit of Smith and, indeed, the Scotland Bill? I think that it is probably a quite good test case now as to how we answer your question. I think that it might be useful if we can maybe get spice to look into that a bit more for us as well, please. I am sure that we can arrange for that to happen. I will come back to you on the board on later. I think that Linda had a supplementary. Do you want to deal with the no detriment stuff for your work? Whatever you like, but I had a particular supplementary on what Jim just said. It was about the employability services. The budget has been slashed, as you have been talking about. You mentioned Smith and the spirit of Smith. It just strikes me that the difference between what was agreed should be transferred and what is now likely to be transferred in terms of funding is massive. It would seem to me to be not allowing Scotland to take on the role that was agreed in the Smith commission, it should take on even in the limited agreement by the UK Government as to what should be passed over the employability services. I wonder if you picked anything up, Jim, heard anything about how this was going to be managed. I know that negotiations must be on-going, but it would seem to me to use the well-hackneyed word that there is a great detriment to the Scottish Government and, indeed, to people in Scotland as to what is now being transferred as opposed to what it was agreed would be in the spirit of the Smith commission. David and I were both at a resolution foundation event yesterday on trends in the Scottish Labour market, where the cabinet secretary, for fair work, was speaking on that very issue. I am sure that it will be debated in the Scottish Parliament very soon. The Smith commission and the Scotland bill, as amended by the Lords, is very clear about which programmes and why. It does not, of course, talk about resource levels. That depends on the point at which you cash out the settlement and try to index it. It just so happens that there is a major reduction in the budget attached. It is possible that that will happen in future with other budget lines. I think that it is important to keep hold of David's point about the intended eligible population. Work programme is going to be blended into a much broader health and employability support programme through Job Centre Plus, as I understand it. There may be a way of trying to calculate what Scotland's broad share of support for the eligible population is. The figure might be a bit bigger than the figure that was simply announced in the autumn statement. I think that it is important to make the point that this is not about views on the merits or otherwise of the work programme and work choice. The issue that is relevant to this discussion is that high-performing welfare to work programmes, based on international evidence, can help to improve employment rates and work incentives, especially where the employment labour market is improving. That, in turn, can help with your income tax revenues and other consequences. It is not simply about what is the budget to be devolved. It is also about what is the knock-on effect for parts of Scotland with less strong labour market areas. What can the offer therefore be? There is also a consequence in Scotland for how we rationalise and improve the effectiveness of existing employability programmes that are run locally, which are for different client groups. Nonetheless, we have to have a sharper focus on our existing powers, our existing budgets and how effectively they are being deployed. That is a separate but important consequence of the decision. We have just asked for a bit of clarification. How difficult will that be with the changes that have been made? We have always talked about that it is not about the powers, it is about what you can do with them. Where I come from has always been that we have not got enough power to be able to do what we want. It would seem to me that, even since the agreement and the spirit of the Smith commission that we are reducing the powers in real terms about what you can actually do. Already, before transfer, we were having potential budget cuts in what was agreed at that point. I know that it is not all about money, but it makes a difference as to what you can achieve. If you come at this from the perspective of what are the outcomes for the intended participants and beneficiaries, we are looking at something like 17 to 20 per cent of the original budget prior to the autumn statement. We could probably turn that into 25, maybe even 30 per cent of value for participants by maximising what is devolved and maximising the effectiveness of local employability programmes in Scotland. As an example, Fife has just conducted a major review of its employability services and, as a result, it has substantially improved job outcomes over the past four years. I am not going to say that we can't achieve much more, but this is a ballpark figure. We could probably get back to about 25 to 30 per cent of value and outcome for intended participants, but that is clearly one-third of what we thought might be coming. The decision will be for Scotland, but I would imagine that it will be about a much more targeted offer to many fewer people, perhaps even geographically more specific. There might be an issue here about City Deal in Glasgow and Clyde Valley, where a different kind of deal can be offered to those eight local authorities, but that is unsatisfactory because in every single part of Scotland we will find long-term unemployed people and people who are stuck in low-pay and in-work poverty who could benefit from a much more ambitious type of offer. Our ability to do that will certainly, over the next five years, be constrained, sometimes constrained. I think that we will come back to that, because of all the things that we are hearing about, how we are going to manage some of the fluctuations through this, which comes down to the borrowing questions that Stuart McMillan wanted to ask around. We will go there first and come back to the detriment issue of individuals. In written evidence provided to the Finance Committee, David Phillips pointed out that the current borrowing is only available to the Scottish Government at present to cover forecast errors of tax revenues. With the new welfare powers, they clearly have to be increased. What type of regime should set limits, if any, on borrowing that might be spent on welfare payments? Borrowing in relation to welfare is not something that is normally thought to be overall good macro policy. Borrowing in relation to invest is the more... Now, spending on welfare might be a form of investment, difficult to figure that out. Certainly, the so-called golden rule around macro fiscal policy rules is around you should be borrowing to invest in the capital stock of the country rather than for current revenues, because there is a danger of ultimately running yourself into borrowing problems. In relation to the welfare powers, the ones that Scotland has got are not really likely to be subject to substantial cyclical fluctuation. They are mostly fairly stable sums of money, so it should be reasonably easy to forecast them on a year-to-year basis. Borrowing associated with forecast errors in relation to welfare needn't, it seems to me, be that large, precisely because benefits like job seekers allowance, perhaps housing benefit of the margins, because they are not being devolved. Whatever we think about that decision, the consequences, as David said, are much smaller in the Scottish budget looking ahead. There will be decisions about the amount of borrowing and bond issuing, which is an important part of that, too. We would say that the debate around that should be much more about investing in the childcare infrastructure and affordable housing supply, which are the drivers of some social security demand. As far as possible, we want Scotland to be taking control of those longer-term drivers and investing in productive activity, which gives us some degree of financial control. Capital spending is where the major draw-up on borrowing and bond issuing should go, but we could be more creative about how we define productive investment. I think that Chelky on housing are two good examples of how we can do that. They fit into the kind of prevention agenda, and you can make the case that you are borrowing now to prevent extra spending in the future. Of course, one of the key issues is how you borrow. In what form do you borrow? Do you borrow on the markets? Then you have to be concerned about what kind of premium the markets might add, depending on what you say you are borrowing for. Or are you borrowing from the UK Government, or, in effect, through UK Government channels, in which case it gets subsumed into a very large chunk of borrowing that is happening each year? Again, that will become clear when the fiscal framework is published. I thought that that was an interesting answer, particularly in relation to Chelky. I had not really considered that to be one of those in relation to the borrowing. Do you want to kick off on issues to do with individual detriment and then come to Robin for fiscal consolidation issues? I am sure that there will be some other questions about that. It is just the general issue, and it is one that has been kicking around since the days of the Smith commission through the welfare reform committee. It is the question of the individual detriment. Again, that is what you want to do with the powers that you have and the ability to give extra income to people, whether it be by new benefits or top-up benefits. I am not yet convinced that we have bottomed out the issue of not giving with one hand and then it being taken away with the other, because some benefits are means-tested. If we are talking about UK Government policy on sanctions and the ability of the Scottish Government to be able to give grant to alleviate hardship there, does that mean that it is taken back in some other way? Can we assure that there is no individual detriment to households where the Government in Scotland would want to either top-up existing or give a different kind of benefit? I hope and believe that we have made progress with the last stages of the Scotland bill in terms of both Governments being clearer about what that ought to mean. If you take the example that we talked about earlier, if Scotland were to choose to find a way of retaining the value of the work allowance, for example, within universal credit, that is good for work incentives, so it has potentially positive revenue consequences. If we were to do that, it would make absolutely no sense economically, physically and in labour market terms for that benefit to be somehow eroded in other decisions within universal credit. We should pay attention to those shared spaces, so housing flexibilities is a good example where different areas of Scotland may even have different rates of LHA variation. Those concurrent or shared spaces versus fully devolved areas, but in both cases it is really important that the net benefit is retained because, if you think about it, Scotland would be choosing to invest more or forgo more revenue in certain areas. Back to the incentives point, there has to then be an incentive around being able to capture the benefits of that expenditure, for example, in higher employment rates. It is in neither Government's interest to stand in the way of that. Again, the point that I made right at the start, because we are in uncharted territory, it may be quite important that we have designed good enough disputes resolution mechanisms that make sure that the spirit of that individual no detriment is properly adhered to in future. As Jim said, we do not really know what the consequences of the relative moves might be and it would be good to have some way of revisiting the situation five years hence or whatever. To see to what extent the objectives of the Scottish Government and the objectives of the UK Government have been satisfied within the settlement as currently agreed, and then what kind of variations might be mutually agreed to get the kinds of improved outcomes in Scotland that you want. You mentioned, right at the beginning, about the divide line between major detriment and minor detriment. Some of the major stuff is obviously dealt with, and also stuff that can be dealt with at the point that happens can be easier to calculate the figure then, but the second tier detriment or no detriment issues are much more difficult. What is the divide line effectively between major detriment and minor detriment? At what point is a cash figure you have in mind or how do you define how it falls into one of those two areas? What should we ignore and what should we not ignore? There are two broad tests. One is the point about proportionality and it is about impact on global budgets. The amount of money at stake, plus or minus, as a result of decisions globally or for particular groups in the population who, you know, the population exposure might be small but the impact for those affected might be very large. That is one route into proportionality. The other thing to look at over time is where we can get closer to satisfying ourselves that there is a causal connection between a choice made and the impact upon revenues and budgets. That would suggest that we are crossing the threshold or reaching that level of some kind of measure of detriment. The only way to get into understanding causality is to really try and go after the likely candidates. You cannot do that for everything. I would suggest that we need to look at suspected major causal effects on both tax and welfare sites and have an intergovernmental focus on trying to understand what are the causal effects. This is art, not science, and it certainly needs to have enough independent weighing up of the evidence. However, this is building the evidence as we go. There is no rulebook that tells us where we are going to get to and it has to be entered into a spirit of negotiation, goodwill as far as possible, but ultimately dispute resolution is necessary. The more light of transparency that we can shed upon the issues, the better. We will need the Scottish Fiscal Commission, the OBR, IFS and others to be in this space. We need to bring good expertise to bear, but it is very unpredictable where we get to. I think that you can only do that in a few areas over the next five years to try and really get at where the big flows might be. Can I take you back 15 years to the free personal care legislation? There were many newspaper headlines around people flooding over the border to take advantage of the much better care provision in Scotland. We did not know what would happen. Now nobody talks about it at all because, as far as we are aware, it did not happen. We did not know beforehand, but we have learned as time has gone on that that is not a big issue. Once things do become apparently a big issue, there has to be a mechanism, transparent mechanism, to resolve the issue. Two questions. First of all, when you talked about proportionality, we should base it effectively on proportionality. Then you talked about going after the big stuff. I paraphrase slightly, but obviously that is what you are saying. If we are using proportionality as the test, the impact on an individual could be extreme, although the relative amount of money would be tiny. Can you explain and expand a little bit what you mean by proportionality? The general point that was being made was about setting up independent mechanisms to negotiate and judge the negotiation between the two Governments on the big stuff, how you define what the impact has been. My question is on who guards the guards. Effectively, who checks what is the resolution process to disputes? Where do you go with that once you have entered into that kind of process? I will have a go at the first question on proportionality. As well as potential global impact on budgets, we want to have some handle on population groups that look as if they are particularly exposed to detrimental consequences of either tax or welfare decisions. There will be reasonably good evidence to guide us as to the likely groups that might be most exposed. If you think about changes on tax that either increase or decrease net household income, for some groups in the population could have a major effect on universal credit entitlements, or take a Scottish example, councillor tax reduction entitlement, or whatever replaces it in future. The crossover points where people are affected or not could be quite fine, so the numbers affected could be quite large or quite small. However, we have to have some handle on the likely exposure to risk for those affected groups. Focusing on whether we think that the major proportionate impacts are likely to be but, by minor detriment, I am not suggesting that it does not matter. I am just suggesting that there may be areas where the net effects of cross-bord decisions may net off to close to zero and we do not need to worry about them long to too much. We keep an eye on them but we can put more of our effort into understanding the bigger flaws. It is impossible to completely micromanage the situation in that there will always be some losers as well as winners if you have a broadly fixed budget, but that is the case now. I do not think that we have really discussed mechanisms enough around dispute resolution. There should really be a set of agreed rules that cannot be changed without both parliaments agreeing on how disputes are going to be resolved. There has also got to be some kind of third party, which is not the Treasury, which has never been a third party, which can take a view on a lot of the issues that we have been discussing, the tax issues, the welfare issues and so on, and has the respect of both Governments and the electorates. That is a tough call. The OBR is supposed to fit into that role. It has made its way. I do not think that it is fully viewed fairly or unfairly how it is viewed. It may not be fully thought to be independent, but there is not anything in train at the minute that seems to me to be around to set up the rules and to set up the institution. We will have to see that in part of the fiscal framework, but we will have to see in three weeks or so if something like that could be agreed. Stability, it seems to me, would require something that is transparent and accepted by both sides. Perhaps it is a very obvious point to make in this company, but, as far as possible, you would want to reach, first of all, a political settlement through the joint ministerial committee. Only if at that level we did not reach good enough agreement would you go to a more technical form of dispute resolution where both Governments would appoint, presumably, representatives to take that forward. As far as possible, those are fundamentally political decisions about resource allocation, about economic management. We need to let accountability take its course through politics. If possible and if not, we need to have a good enough backstop. I wonder on which one of you should be applying for the job for this independent position. Because, whether it is the tax side or the welfare side, there is no arbitration mechanism effectively. If it is not in the fiscal framework, are you encouraging us as a committee to make a recommendation that something around that arbitration on both the taxation side and the welfare side should exist? As a supplementary point to that, can you know the role in scrutinising welfare as well as tax forecasts? The OBR does some forecasting of welfare spend relative to DWP who also conduct forecasts. For that paper, I used the OBR forecast of welfare spend and not the DWP ones, because the DWP tends to be very optimistic. It is a different type of forecasting, all the same, from macroeconomic modelling on the one hand to what is going to be the change in eligibility for a particular part of the population in relation to welfare benefits. Those are two quite different strands of analysis. You should look to setting up a competition for a body to do the forecasting on the welfare side, rather than having a monolith that tries to do everything. I am not sure. There is a lot of work done by independent bodies in London around welfare in particular and health spending as well as policy. There are mechanisms for building up expertise, and it could be brought within the framework of the Fiscal Commission, but it is a different kind of work. I was trained as a geographer, not an economist. I will be a little bit critical of our economics fraternity in the sense that forecasting subjects to even the best forecasting subjects is such large margins of error, especially at times of such uncertainty in public finances. It is important that we get as good forecasting as possible and future finance secretaries and chancellors and parliaments will want to be assured that the information is as good as possible. However, the important point that came out last session is how good the out-term measures and the reconciliation are at the end of financial year or even if it is over a three-year cycle, getting accurate information on the actual position on spending and the earlier point about what underlying entitlement is. I am not losing that point about take-up. We are improving our measurement capacity and comparing that backwards to our forecasting. Having the two in tandem I think is what is going to improve the position on both sides of the border. So, convener, it means that we have to have a mechanism to do that in Scotland, even before we come to any potential arbitration between our views and the views in London. I mean, I think that there is an interesting question about the relationship between, let's say, the Scottish Fiscal Commission and the OBR. To what extent, for example, the Scottish Fiscal Commission and its forecast for Scottish income tax takes on board the macro forecast for the UK as a whole, because they are undoubtedly interdependent that the OBR makes. Again, as far as I know, whether that will even be agreed in the fiscal framework, I do not know. In a sense, you probably do not want two governmental bodies competing with each other to forecast the UK economy. That is probably not a brilliant idea, but I do not know what the arrangement is going to be. You do it more than fiscal consolidation. I do indeed. You painted a pretty bleak picture of the available welfare support in Scotland's ability to be flexible over the next five years. The Command Paper Scotland, the United Kingdom during settlement, states that, therefore, the fiscal framework must require Scotland to contribute proportionally to the fiscal consolidation at the pace set out by the UK Government across devolved and reserved areas. Professor Michael Keating has stated that it appears to go beyond the requirement that any extra expenditure in Scotland be financed by Scottish revenues, which is already covered by the balanced budget requirement. What are your views on that? Well, I am not convinced, I agree, because increased spending in Scotland financed by increased taxation in Scotland does not add to the UK's borrowing. If you think about the fiscal framework being A, a deficit target and B, a debt target and the deficit is intended to be eliminated by 1920, then, if Scotland does increased spending that is tax financed, I would have thought that that does not undermine the UK's fiscal targets. I cannot add to that. I suspect that it is 2020, not 1920, as you have suggested. Sorry, I meant fiscal year 2019-20. I left out the hyphen. I am reasonably okay with that. In the written submission to the Welfare Reform Committee, you noted that a number of benefits being transferred to Scotland fall within the welfare cap. However, the DWP will not be required to account for the spending in relation to the welfare cap. That is made clear in an enduring settlement document. What is not clear is whether the responsibility will be transferred to the Scottish First Minister. If the responsibility is transferred from the DWP to the Scottish Government, will this impose significant constraints on Scottish policy autonomy? If extra spending is funded by extra taxes raised in Scotland, would the welfare cap still apply? That is taking it on from the point that you just made. That is one part that is not covered by my previous answer and would impose some responsibility in relation to overall fiscal targets. I suppose that you could argue that the welfare cap is a means to achieve the ultimate goal, which is the deficit and the debt targets. A transfer of that responsibility to Scotland would have an impact on Scotland. If you look at benefits being devolved currently covered by the welfare cap, so it tends allowance, cares allowance, DLA PIP, cold weather payments and winter fuel payments, they are not about how you are not showing up a failing economy by spending money in these areas. Social investments to largely older people, largely disabled people or people who live in very cold parts of the country. I suggest that they have a very tenuous connection to the chance of stewardship of the UK economy overall. I would also say that, both on welfare cap and on the household benefits cap, we are sceptical not from an ideological perspective, we are sceptical about the effectiveness and we are sceptical about some of the perverse incentives that they build into the system. Of course, controlling spending in these areas is important, but the best way to do that is by tackling the underlying drivers. That is about unemployment, childcare and housing supply, which are nothing to do with the bulk of welfare cap. I cannot answer whether it will happen, but our view would be that it should not. Thank you. That brings us to an end of what I think has been a fascinating evidence session taken. Given that there is a lot of food for thought, I would probably beg as many questions as it has answered in my own head if the truth be told. I thank both witnesses for coming along today, it has been very helpful and thank you for giving us your time. I want to move straight into agenda item 2, which is an opportunity for the committee to consider recent and forthcoming work by other parliamentary committees on issues raised by the provisions in the Scotland Bill. There are attached to the paper as a report from the Local Government Regeneration Committee on fixed odds betting terminals, a copy of the letter of the convener of standards and procedures, public appointments and committee on supermajorities and powers to amend the Scotland Act. Additionally, there is information that the infrastructure capital investment committee is planning an evidence session on off-com. Finally, we know that we are going to be getting some correspondence from the rule of fair climate change environment committee on its further work on the transfer scheme for the Crown estate. I do not know if the member has any comments, but what I would suggest is that today we note those and what obviously in terms of content, we need to use those as part of our consideration of a report in due course. No, I gave you that, convener, but I would just like to put it on the record that the work of the Local Government Regeneration Committee on the fixed odds betting terminals piece of work that undertook, I thought was excellent and it is similar to all the issues and I want to commend them for it on the record. Okay, that's on the record. Any other points at this stage or are we happy to note and consider as part of a report? Okay, that's agreed. Before moving into private session, I should add that the committee will meet again next week when we'll take evidence from a panel witness on the issue of post-study work visas and discuss the report on a recent fact-finding visit to Spain. I now move this meeting of the committee into private session.