Official Report

Good morning and welcome to the Finance and Constitution Committee’s ninth meeting in 2019. As usual, I ask everyone to ensure that their mobile phones do not interfere with proceedings.

Our only agenda item is to take evidence in a round-table format on earnings in Scotland. We will take evidence from Anna Ritchie Allan, executive director of Close the Gap; John Gallacher, Scottish organiser at Unison; Torsten Bell, director of the Resolution Foundation; Professor David Bell, professor of economics at the University of Stirling; Helen Martin, assistant general secretary at the Scottish Trades Union Congress; Hazel Brown, leader of exceptional service (quality) at Cornerstone; and Russell Gunson, director of the Institute for Public Policy Research Scotland.

I warmly welcome all our witnesses and thank them for coming along to help with our proceedings. Those who have been involved in a round table before will know that the format is intended to create as much free-flowing discussion as we can achieve. I ask anyone who wants to contribute to let me know or to catch the eye of the clerks, so that we can get you in. The discussion will be based loosely around four themes, and an MSP will kick off questions on each theme to get us going. Witnesses should feel free to contribute at any stage.

James Kelly will get us under way with questions on how public sector pay compares with private sector pay. I am the first to recognise that it is inevitable that comments will cut across from one area to another.

I thank everyone for coming along to provide their expertise and experience. The first theme that we are interested in exploring is the difference between pay levels in the public and private sectors. One trend that has been identified is that public sector pay has a higher premium than private sector pay, and that applies more in Scotland. We are initially interested in exploring what the differences are between the public sector and the private sector when it comes to what drives wage levels.

Who would like to kick off with their first thoughts on that question? I see lots of people looking at me and being quiet.

My opening salvo is that we should not take it from the figures that workers in the public sector are doing well on their wages. Real-terms earnings are at about the 2008 level. Until last year, when the Scottish Government lifted the 1 per cent cap on cost-of-living negotiations, wages were flatlining. Some sectors have fared better than others, but some sectors in Scotland still labour under the 1 per cent ceiling for United Kingdom departments and civil servants, although they are not Unison’s concern. Social care in the third sector is also struggling with wages.

We should not say that all in the garden is rosy for the public sector. The year 2018 was okay—I would not describe it as more than that—but it did not address catching up.

The public sector is not a homogeneous unit. Pay for similar jobs varies wildly across different local authority areas, universities and colleges. We should not assume that workers who do the same job in different parts of Scotland are on the same wages.

I sit on the board of the Coalition of Care and Support Providers in Scotland so, although I speak from Cornerstone’s perspective, I also have that experience. James Kelly asked about earnings but, as an employer, we cannot compete at all with the public and private sectors on terms and conditions—we have no chance of offering similar pensions and sickness provisions.

When local authorities started to outsource social care about 15 years ago, some research was done and, at that time, as a third sector social care provider, we were able to offer similar terms and conditions. However, that ability has been eroded over the past 15 years, so we can no longer compete. If, say, North Lanarkshire Council advertises posts for carers, a glut of our people will go for them, because we cannot offer the same terms. Even though, as a provider, we try to offer hourly rates that are better than the Scottish living wage, we cannot compete when it comes to the whole package.

The differences in average pay reflect the differences in the work that is done in the public and private sectors. Typically, the work that is done in the public sector is more skilled, although there is a spread across the distribution of the workforce. The movement of the gap between the public and private sectors has been of interest. The private sector suffered more at the start of the recession, but the subsequent pay restraint in the public sector has narrowed that gap.

As Hazel Brown said, earnings are only part of the story, and the whole package of terms and conditions matters. With regard to pensions, dare I say it, workers in the public sector tend to be better provided for than those in the private sector. In the round, public sector packages are typically better, but that neglects the fact that there are people in the public sector who are not that well paid and who struggle. In straight comparisons, we find that the gap between the rich and the poor in the private sector is bigger than the gap between the rich and the poor in the public sector.

I do not want to replicate what Professor Bell said about the big picture. The difference in the picture across the UK is that the public sector premium looks bigger in Scotland. That is partly driven by the fact that there has been slightly better pay growth in Scotland over the past few years, but it is largely driven by the lower private sector pay in Scotland. Therefore, the premium does not reflect higher public sector pay; it relates to what is going on in the private sector.

The fact that the public sector makes up a slightly bigger part of the Scottish economy overall is very important for the big picture on pay in Scotland, including the reasonably fast catch-up and overtaking on some measures during the 2000s, which we may come on to discuss. That was partly to do with Scotland having a larger public sector that saw larger pay rises in the run-up to the financial crisis, especially in the last few years leading into the financial crisis. Although we are all focused on the relative gap, what drives it and the different time periods are important, rather than just how big the premium looks in raw terms.

I add that there are high levels of low pay in the private sector in Scotland. In some industries, collective bargaining coverage is very low. For example, in hospitality, agriculture and some parts of construction, particularly where there are tier 2 and tier 3 workers, there are people whose pay has been consistently pushed down over the years and who therefore suffer from low wages and insecure work. That is partly to do with the low level of collective bargaining coverage and the inability of unions to bargain for those workers, for various reasons in different sectors.

It is important to recognise that we have had a large living wage campaign in Scotland, which has been very successful as regards the number of companies that have signed up to be living wage accredited. We probably have the highest number of such companies in the UK, yet that has not moved the measure of how many workers are suffering from pay that is below the living wage—that figure is static at around 20 per cent. Therefore, some of our focus needs to be on how we raise pay in real terms for the lowest-paid workers and how we improve collective bargaining coverage and the security of work in the private sector in particular.

I would like to make a comment on the gender dimensions of pay practice. Generally, we can say that public sector employers are more likely to have done an equal pay review, which women are more likely to benefit from. However, in the private sector, pay practice that is premised on discretionary pay is widespread, and evidence shows that women are more likely to be disadvantaged by that. The potential exists for discrimination in relation to pay, which sustains lower pay for women. To follow up on Helen Martin’s point about the living wage, two thirds of the workers who earn less than the living wage are women, so there is a clear gender dimension there, too.

About 20 per cent of employees in Scotland are in the public sector, which is a big chunk of those in employment. We need to look directly at the levels of pay in the public sector, but we are also interested in what can be done in the public sector that can spill over into the private sector, for example by competing in local regional labour markets or through procurement and the softer power that the public sector is likely to have over those who are paid at low levels in the private sector. We have a direct policy interest in what pay looks like in the public sector, but we are equally interested in what can be done to influence the private sector through setting pay in the public sector.

Helen Martin said that there is better take-up of the living wage in Scotland. Is that mostly driven by the public sector, as Russell Gunson hinted at? Are companies in the private sector coat-tailing—for want of a better word—on what the public sector is doing, or is it the other way round?

There are more than 1,000 living wage accredited employers in Scotland now, and they come from a variety of sectors. However, they tend to be companies that do not have high numbers of low-paid workers in their workforce, so it is not the case that high levels of retail employers, for example, are signing up to be living wage accredited. In that respect, the living wage is making a difference around the margins, but the overall picture of workers who are paid below the living wage in Scotland is stuck at around 20 per cent. It is a puzzle that although there is a high level of accreditation, that is not having the impact that we would like it to have.

We set the living wage rate. Scotland has a slightly higher level of living wage employers, but when it comes to the labour market as a whole, that is not a factor. The living wage campaign is a really important campaign, which we support. We are a member and we sign up to it, but it is not yet the case that enough people are affected by the living wage for it to be something that matters for Scotland’s labour market as a whole, much as you might want that to be the case. There has been a good campaign for the living wage in Scotland, which is public sector led, as it is in some other parts of the UK. However, although it is important for the people who are affected by it—that is why we spend a lot of time calculating it correctly—it does not change the average earnings rate. The numbers involved are too small.

That takes us back to Russell Gunson’s point, which was that using the public sector as an exemplar has not been as successful as we might have expected.

Wales and Northern Ireland, which also have a higher proportion of employees in the public sector, have far higher numbers of employees who are paid below the living wage. Something different is happening in Scotland. Around 19 per cent of employees in Scotland are paid below the living wage. In Wales, the figure is 26 per cent and, in Northern Ireland, it is closer to 30 per cent. Those are quite significant differences, which suggests that there must be factors at play in Scotland’s pay structure other than the dominance of the public sector.

It is still important that public bodies get living wage accreditation because, under the terms of that scheme, if they outsource contracts, the living wage rate needs to apply to those contracted-out services, although not necessarily immediately—sometimes, there is a time lag of a couple of years. A public body that is living wage accredited must apply the living wage to contracted-out services at some agreed time. That is important, because it would be very difficult to achieve that by other means, particularly when some of the companies concerned do not recognise trade unions and there is no collective bargaining mechanism. In some cases, achieving £9 an hour would probably involve industrial action.

Obviously, we would prefer public bodies not to outsource at all, but it is important that as many public bodies as possible get living wage accreditation, because that means that, when outsourcing occurs, a discussion has to take place about the wage rates of the privatised provider.

I was interested in Russell Gunson’s comments about the influence that public pay could have on the wider economy. If the kind of sectors that Helen Martin was talking about, such as retail and construction, are the ones that cause most concern, they might also be the sectors where what the public sector offers in pay and remuneration has less influence on the private sector. Are there ways to maximise that influence and leadership role? If so, what are they?

On public sector pay in its own right, just a few months ago when the Cabinet Secretary for Finance, Economy and Fair Work announced the public sector pay policy for the current financial year, he said that it

“continues the journey of restoration of public sector pay.”—[Official Report, 12 December 2018; c 35.]

That was a description of a pay settlement that almost reached inflation at the low end and did not reach inflation for other people. Is there a journey towards restoration of the value of public sector pay? Do people have any indication that they expect that journey to happen or to continue?

I will pick up on Russell Gunson’s point and say that—Hazel Brown might have been going to say this—the care sector, which is probably the fastest growing sector in the economy, is in a complete bind because, on one hand, the living wage is pushing up costs and, on the other hand, there is the squeeze of the local government contracts and what local authorities are prepared to pay per week for someone who receives care in a care home. As a result, we are not seeing any expansion of the sector, because it is fundamentally unprofitable. Although we face a challenging demography over the next couple of decades, we are not seeing growth in care provision, largely because of the squeeze that is affecting the sector.

David Bell said most of what I was going to say. The squeeze is the issue. The application of the Scottish living wage in the social care sector—one of the lowest-paid sectors—is great and we welcome it, but organisations like us that operate across Scotland struggle to implement it, because all the local authorities calculate the living wage differently.

The concern is that although, for the past few years, a particular amount of ring-fenced money has come to the sector to support the Scottish living wage, it increasingly appears that, in the future, our commissioning partners will expect us to meet that as part of business as usual. That is yet another squeeze that the sector cannot cope with.

We cannot talk about care without talking about women, given that they comprise the majority of the workforce. That applies to the early years workforce, too. Care work is so low paid because of the economic undervaluation of the work itself, which is inherently gendered. The literature says that it is quite difficult to challenge that, but one way of doing so that has been suggested is using state wage-setting powers. That has been done piecemeal through the Scottish Government’s commitment to pay the living wage to early years childcare workers who are delivering the funded entitlement. However, that creates some challenges, because it applies only to the funded entitlement and so does not address the wider problem.

If we are to tackle the undervaluation of the type of work that is done by women, there needs to be a much more strategic investment approach that identifies care as a growth sector and includes it in the Scottish Government’s economic strategy.

Patrick Harvie asked two broad questions. The first question was whether we were on a journey to catch up on public sector pay. There is a long way to go—if we are even on that journey. Across the whole economy, real wages are still lower than they were before the crash. We have never seen anything like this in modern history—it is 10 years and counting. There would have to be a long catch-up period, with above-inflation pay increases across both the public and private sectors.

The UK is in unprecedented times; the projections covering the recent past—and, in the short term, the coming years—are that Scotland will underperform against even the UK’s very poor wage growth. We need to get inclusive growth and the economy going, so that pay rises across the public and private sectors can be above inflation to catch up.

On how to maximise the influence of the public sector over sectors that are more distant from public spending, care and childcare are good examples of where there is a close relationship and the hope that the public sector could do more to boost pay directly. We should not forget that it is not about only the pay floor, although the living wage is important; career progression and job quality are as important, if not more so.

For the big lower-pay sectors in Scotland, such as retail, we must look beyond pay in the public sector. We must look at how we can increase collective bargaining, take a social partnership approach across the economy, drive productivity increases in those sectors rather than only in the growth sectors, and, most of all, get those productivity increases into the pay packets of workers at the low end. Having pay rises at the low end that outstrip not only inflation but pay rises at the higher end is how we deliver reality around all the good words about inclusive growth.

On low-paid workers, the big picture is of very fast rises in the national living wage—the minimum wage for workers over 25. That is the single largest thing that is happening to hourly pay distribution. Across the UK as a whole, that is a big deal. It is a slightly smaller deal in Scotland than it is in most of the UK, for the good reason that pay growth in the 2000s in Scotland was broadly based over the bottom half of income distribution. Scotland has a less unequal pay distribution because it does not have as much at the very top. It has about the same as the rest of the UK, minus London and the south-east. If that area is treated as a freak show, Scotland is broadly in line with the rest of the UK.

The bottom of the income distribution is better in Scotland: it is nearer to the typical, and the national living wage has a smaller bite. The national living wage lifts the bottom, but fewer people are on the bottom in Scotland. That is good news in some ways, because it reflects better pay distribution, but it is bad news in other ways, because lower-paid workers get less of an increase in Scotland than they get in some other parts of the UK. In that way, Scotland looks a bit more like London, where the national living wage has less bite because there are fewer workers on the legal minimum. However, there are obviously housing costs galore there, so it is not really doing them any good anyway.

Patrick Harvie’s initial question was how we can push out influence to industries such as construction, and we have done quite a bit of thinking on that. We were pleased to see collective bargaining coverage included as an indicator in the effective voice element of the national performance framework. We have had discussions with civil servants about how to move the collective bargaining coverage indicator, thinking about the construction, hospitality, social care and childcare sectors. Each of those sectors works differently, because they have different starting points.

There are elements of construction where the workforce is well organised and national agreements already exist in which terms and conditions have been bargained. The issue is that those agreements are not always enforced and not all workers get access to the terms and conditions. We must make sure that all workers have access to a union and that the union has the opportunity to organise on and get access to the sites. The construction sector is scarred by blacklisting and ruthless behaviour by employers. We must use the weight of public sector commissioning to ensure that such things cannot happen.

Unite the union has produced a construction sector charter that we want to be incorporated into public sector contracts. We are also looking at whether we can put in place helpline and compliance measures, so that workers can get in touch and report poor practice.

We also need to ensure both that workers understand the collective agreements that are in place and that those agreements are honoured right through the supply chain of public sector contracts. The initial company might be okay, but, going down through the supply chain, things get increasingly worse.

The use of umbrella companies increases pressure on pay and terms and conditions. The focus on collective bargaining is really important; as Russell Gunson says, it is about more than pay—it is about terms and conditions, job security and how much workers feel that their jobs benefit them in the round.

This has been a good and useful opening discussion. We will go to John Gallacher and then Torsten Bell, and then we will move on.

I will be quick. Helen Martin has covered a few points; we say “Hear, hear” to sectoral bargaining in the early years and adult social care sectors.

The progress on the living wage is welcome, but it masks a basket of jobs for which the living wage has been accepted as the rate for the job although it is not necessarily the right rate for the job. I mean no disrespect to shop workers, but if stacking shelves attracts a living wage rate, that means that it is not the appropriate rate for delivering the 0 to 5 curriculum in an early years setting or adult social care to vulnerable elderly people in the community. It masks a can of worms and we should not assume that, because the living wage is applicable, that is a good thing.

To answer Patrick Harvie’s question directly, there is no real “journey of restoration”, nor will there be an opportunity to negotiate pay in Scotland in the public sector for a couple of years. Local government, health and the college sector have just signed three-year deals. It will be 2020 before we next put in claims that might have a restoration element.

I have some examples. If an early years worker in the local authority sector who was on £20,228 in 2014 had had wage increases linked to the retail prices index, they would be on £21,929. However, they are sitting on £21,392—they are £500 or £600 behind. A registered nurse on a middle-ranking grade should be on £26,358 with RPI increases, but they are actually on £24,500. “Theft” is a strong word, but there has been a significant and real theft of wages from public sector workers over the past few years that few people expect to get back. They have suffered the pain and impact of that on their personal budgets—which they resent, to be honest, because their jobs have not got any easier.

I will make just one point, given where the conversation has gone. On what the public sector can do to raise wages, the state can make lots of important, direct micro-interventions, where it either procures the service or uses its leadership role.

However, the biggest thing that is driving slow wage growth is weak productivity growth. It is not changes in union bargaining power. Trade union density has carried on falling at roughly the same rate that it was falling at before the crisis, although it may be falling at a slightly slower rate now. The levels are a bit higher in Scotland, but the big picture is the same. The increasing monopsony of big employers is not what is driving weak wage growth.

The problem right now is that we are not seeing productivity growth. With regard to what caused that, there was a big financial crisis, then higher unemployment in the phase afterwards. However, the problem now is lack of growth. If you want to boost wages across the economy, you need productivity growth and faster growth. Although Scotland’s growth in the past 18 months to two years is broadly back in line with UK averages, that is because UK averages declined again over the past two years, as you might have noticed. Now everybody is doing a bit rubbish. That is the most important thing.

I think that that point gets lost in a lot of these discussions. We focus on what we can do for 10 people here or 10 people there—rightly, because such policy feels easier. However, that will not solve the problem. The reason why wages are below where they were 10 years ago and why wages look as though they will grow more slowly over the next five years than they have ever done is weak productivity growth—first, last and always.

The committee has been over that ground quite a few times. We have not found the answers to some of that yet, but perhaps we will get enlightenment today.

That was a very helpful and useful opening discussion and takes us into our next area. Angela Constance will ask about how pay in Scotland compares to pay in the rest of the UK and the implications that that has for our budget and the fiscal framework.

I often get frustrated at what can appear to be our obsession with comparing Scotland to the rest of the UK, sometimes at the expense of missing the wider point or the bigger picture, whether that is at a European Union or international level. Nonetheless, for this committee, in the context of the fiscal framework and, therefore, Scotland’s budget, how revenues per capita grow in Scotland in comparison to the rest of the UK is a fundamentally important question. There is an obvious relationship between earnings growth and revenues growth.

I am specifically interested in the fact that the convergence between earnings in Scotland and those in the UK from 2013 to 2016 appears to have stalled and plateaued. Worryingly, more recently, the 2017-18 figures show that average wages in Scotland grew less quickly than in the UK. I am interested to know people’s views on why that is and what the consequences are. Is it a coincidence that that has occurred at around the same time as Brexit uncertainty? What can and should the Scottish Government do to assist?

As you outlined, that is a crucial point. Given how dominant income tax now is to the Scottish Parliament’s budget, earnings growth will be at the heart of any differential between what we would have had before the devolution of income tax and what we will have now. It is already kicking in. As you said, we have seen a negative differential over the past couple of years, and that is projected to be the case for the next couple of years. At the current level, it will probably hit revenue in Scotland by hundreds of millions of pounds each year. That matters, because some of the public sector pay rises that we talked about will be affordable only if we find ways to boost tax revenue. The most sustainable way of doing that is through pay rises in Scotland, underpinned by productivity increases.

What has been happening in Scotland? In essence, Torsten Bell did not quite say “productivity, productivity, productivity”, but productivity is a crucial point that we all talk about. At the economy level, it is quite well established that there is a link between productivity growth and pay. At the sector level and, certainly, at the firm level, how they interact gets more blurry. We have been doing some work with the Scottish Policy Foundation on pay, productivity and inclusive growth. One of the key conclusions from our forthcoming report is that, if we want to deliver inclusive growth in Scotland, we need to focus our productivity policy as much on the everyday parts of the economy as on the growth sectors that the Scottish Government has picked out.

One way of reducing low pay, of course, is to move people from low-paying sectors into higher-paying sectors, but a much more likely way of reducing it, given the size and scale of employment in some lower-pay sectors in Scotland, is to boost pay in those sectors.

What can we do to boost productivity in retail, in tourism, in hospitality and in care? The answers are likely to be found sector by sector, and they are very likely to involve employers, employees and Government working together to make decisions.

For us, the why is productivity, but it is not just that; there is an oil crash, and other things might have happened that are specific to Scotland. It is probably less likely that there is a link to Brexit, given that it has impacted across the UK at the same time, although we can see differential impacts—the further away from London, the bigger the impact might well be. Overall, if we cut through everything else, it is about trying to boost productivity, particularly in low-pay sectors.

My final point is that automation is both a threat and an opportunity in this context. Technological change, or industrial revolution 2.0—whatever we want to call it—has a gendered aspect, in that many of the jobs that are likely to be under threat of change from automation are currently taken by women and are in lower-pay sectors.

The opportunity comes in embracing automation in a way that drives up job quality. I do not think that in Scotland we are quite on that yet. We could do more to look ahead to what automation could bring and to understand, sector by sector, how we can take the benefits in a way that drives the social outcomes that we want to see.

Adam Tomkins wants to burrow down a bit into some of that.

Everything that Russell Gunson said is absolutely fascinating, but I do not understand what the words mean.

That is the big problem.

In particular, I do not understand what boosting productivity in the care sector through increased automation means. Let us take the care sector, given that there is a lot of expertise in that sector around the table today. The report that you mentioned is forthcoming, so we do not yet have it, but what do we mean when we say, “Let’s boost productivity in the care sector”?

You picked what is probably the trickiest sector in that regard. That might have been deliberate—

I am like that. [Laughter.]

I promise to come back to care, but let me first talk about retail. We can see how productivity boosts in retail can, over the long term, lead to pay increases, if interventions are put in place to ensure that gains from a productivity boost in an individual firm or in a sector go into the pay packets of the lower paid. I hope that those words made sense to you.

Productivity, of course, is about getting higher efficiency from the humans—the workers—and the capital. Automation in retail could be a threat, because it could displace lots of low-paid workers, but, equally, it presents an opportunity to drive up the quality of the work that remains in the sector after automation has happened.

Productivity in care is much more tricky to measure. The traditional methods miss a heap of important things. A normal productivity measure would look at input—the number of hours put in and the number of workers—and output, which in that context would probably be the number of people who were seen, so a productivity increase on paper could be awful for the patients or clients who were being seen.

Therefore, we need to tailor how we measure productivity, sector by sector. For care, we absolutely need to look at the quality of care and the experience of patients and workers. From there, we can get a much more tailored measure of productivity for the sector.

In broad terms, across the economy, automation has a role, but I am not sure about its role in care. It depends where technology takes us, but it is probably less about automation and investing in new robots and machines and more about traditional ways of boosting productivity, such as having secure contracts that allow businesses to invest in their workers, as opposed to very insecure contracts, which are, in many ways, potentially self-harming for businesses.

I hope that that has answered your question, to some extent.

That was helpful, thank you.

Interestingly, productivity in care is almost entirely measured in terms of hours of care delivered and the number of people it is delivered to. It is not a measure of quality at all. People have been struggling for quite a few years with how we measure and audit outcomes, because that would give more freedom to the sector to deliver care differently and at different hourly rates than we currently have. It is something that Cornerstone has tried to influence, but our commissioning partners are still stuck on what the hour of care looks like.

On Russell Gunson’s point about automation, we are not going to have robot care assistants or anything like that, but there is an issue about technology replacing human intervention in monitoring people. The changes in that respect over the next few years will be quite radical. People will still need humans to look after them, but they will no longer get 24-hour-a-day care. As that time is getting less and less, the question is how else we can monitor and support people when there is no actual person in the room to deliver that support.

I agree with everything that Hazel Brown has just said, although I should point out that there are some robots involved in social care. It is, of course, care in the loosest possible sense, but in Japan they have a robot seal called Paro that people can cuddle and which I think does some other things. There are some robots in the sector; indeed, we have had them over for events. Because Japan really does not want migration, it has robots doing bits of care work instead. You can have that sort of thing if you want to, but you might well choose not to.

As for retail productivity, which you have asked about, it is growing really fast, because the sector is shrinking. If you want productivity growth, one way to get it is to shrink your sector really fast relative to the population. Retail has probably been the major sector with the fastest productivity growth since the financial crisis. Everyone says that the problem is that low-paying sectors are not seeing any such growth and the high-tech and financial services sectors are. That is true in the long term, but it is not at all true for the recent past. The financial services sector has had zero productivity growth, and the pharmaceuticals sector is seeing zero growth, too. Over the past few years, the big firms have been rubbish. Some of the small sectors have been doing better, and retail is the best example of that.

However, I am slightly worried that we are missing the core question here. What does the shape, volume and nature of earnings growth do to public finances, and what does that mean for Scotland?

And what can we do to help?

I have a really unhelpful answer to that question. There is a real lesson to be learned from last year’s UK income tax figures, which will be important to Scotland. How do we get higher income tax revenues? I think that, apart from putting up taxes, which I will leave to politicians to take decisions on, there are three ways.

The first way is more employment. Scotland had a pretty good year on employment last year, but in relative terms, it is still below where it was before the crisis, when it was well ahead of the UK.

The second is getting a bit more earnings growth, in which respect the past few years have been very bad. My view is that things are bad right now, because of the weak economic growth in 2015-16. That was a sweet spot for the UK as a whole but not for Scotland, and now it is rubbish for everyone.

The third way, which is really important but is understated in our income tax system, is more unequal pay. If you want higher income tax revenues, you need pay growth at the top. It is all to do with the increasingly progressive nature of our income tax system, not, I should say, our tax system. Everyone says that the rich pay all the tax nowadays, but that is not true; what they mean is that the rich pay all the income tax nowadays, and that is because they have all the money. Income tax revenues are overachieving all our expectations at the moment because in the past 18 months to two years earnings growth has flipped from being progressive; it is still progressive on an hourly basis, but it is regressive on a weekly basis.

We then get into problems with the data for Scotland. We just do not have much of it, and you have to jump around, looking at small parts here and there. However, it does not look as though pay growth in Scotland has been as unequal as the UK as a whole over the past few years. It is not just that slightly lower earnings growth will lead to lower income tax revenues; if you continue to have better earnings growth—which is the fairer approach—you will also have weaker income tax revenues. I never hear that being discussed at all, but right now it is probably more important than the other stuff.

Torsten Bell has pretty much covered what I was going to say. However, going back to the point about the care sector, experiments with technology are certainly being carried out, and it will play an important role. It is also important to remember that that might intersect with the Scottish Government’s prevention agenda.

In that sense—and picking up on Torsten Bell’s point about having less of something—I would point out that if we monitor people with technology that is not necessarily intrusive to, for example, prevent falls, which is particularly important for frail elderly women, they will need less care over their lifetime. It is the share of care over their lifetime that matters. That would release resources to be used in other ways. If we managed to get prevention technology working effectively, we would not need as many carers. There are ways in which technology can indirectly affect productivity without our necessarily needing to have robots looking after people.

The point about income distribution is exactly right. Like it or not, we have very unequal income distribution, with the people at the top tending to pay a lot of the income tax. We are, nevertheless, less unequal than the rest of the UK. The difference in average earnings between Scotland and the rest of the UK is not really a great indication of income tax revenues, because people who are being paid the average will not actually pay that much income tax—it is the high earners who contribute a lot of the overall revenue.

It is certainly true that high earners pay high levels of income tax. However, I do not think that that is a reason to prioritise income inequality as a key policy outcome. I would much rather go back to looking at how you move people along the pay spectrum and how low-paid workers become middle-paid and high-paid workers, instead of simply thinking straightforwardly about how we recoup the highest level of tax by increasing the number of high earners.

A short-term way of boosting productivity would be to cut the number of workers and carry on producing the same amount, but that would probably be a short-term road to long-term ruin. A much more sustainable way of increasing productivity would be to invest in the things that help your workers to be more productive, whether that be management, skills or new technologies.

The other point is that productivity growth might not find its way into pay growth, and you might need to intervene to make sure that that happens. Productivity in retail has grown, but that has not necessarily led to pay increases for those at the low end. What can we do there? It is not just about getting productivity growth but about making sure that that finds its way into pay growth.

It is absolutely true that in the short term you could bring in large amounts of income tax revenue by boosting high pay, but we also know that a fairer economy is a stronger economy. Indeed, that underpins the whole inclusive growth agenda. In the short term, you might be able to sustain increasing tax revenues in that way, but in the long term you will undermine the strength of your economy. In the long term, it is much more sustainable to try to boost lower-paid and medium-paid workers’ career progression, because you are much more likely to get more sustainable growth that way, even if that means sacrificing tax revenue increases in the short term to some extent.

Can somebody help me with a quandary? I have heard others talk about automation helping to improve productivity; in effect, it means less labour, or perhaps the labour is doing something different. Automation might be seen as a measure to increase productivity, but it potentially means fewer people in the economy who are paying tax. Is that not a quandary? If it is not, please sort it out for me.

I know that there are one or two economists around the table, but I will have a go at giving an answer, and David Bell can correct me if I am wrong.

In essence, it is a short-term and long-term issue. If you can find productivity increases in the long term, they should recycle back into employment growth, because your economy will grow—and, if you get it right, that will feed into pay growth. In the short term, you might see an impact on employment levels, unless you can manage that well through interventions, such as having a skills system that can broker opportunities between the contracting sectors and the expanding sectors in a way that does not necessarily happen just now. In the long run, however, if managed well, productivity increases—including through automation—should lead to pay increases, employment growth and economic growth.

The lessons from history suggest that previous waves of automation have not damaged employment levels in the UK. There is a strong debate about what effect this latest industrial revolution—the fourth—will have, with some optimistic that other jobs will be found that will emphasise soft skills that cannot easily be replicated. Others argue that we will see a globalisation of services. Over the past 30 years or so, we have seen the globalisation of manufacturing, and there is an argument that many of our services will be put out to international competition. I do not know where that is going, and I do not think that anybody really knows how it will develop, but we have not discussed it enough. Those developments are now really close to happening in certain parts of the world, and Scotland has not really been talking about it.

Angela, is there anything that you want to come back to as far as your initial question is concerned?

Oh, I think that there is something in all of that for everyone.

But you have no other questions. That is what I really meant.

As nobody else wants to contribute, we will move on to the next area, which is changes in the labour market and impacts on earnings. I realise that we have already covered some of that, but I ask Patrick Harvie to start us off.

Several people have already touched on a number of aspects affecting the labour market, particularly automation and casualisation, and I would also highlight the contrast between some of the stats that get trumpeted about highest ever employment or lowest ever unemployment and people’s lived reality in terms of low pay, precarious work and insecure, temperamental or variable income that they cannot rely on.

As for other changes that are coming in, some of the written submissions to the committee talk about the changing balance between full-time and part-time work and whether there is a change in the gendered pattern in that respect. We must also at least acknowledge the possibility—perhaps the strong possibility—that freedom of movement will be choked off, which will impact the labour market, too.

Of those changes, and others, in the labour market, what are the most significant that are already impacting on earnings? How do the people around the table see that going forward? What issues do we need to look at and might we anticipate in relation to changes in the labour market?

That is a great question. Before I look at what is happening right now, I should point out, by way of history, that the big picture for Scottish pay was of continued growth during the 2000s while everybody else slowed. Scotland’s pay catch-up is a function of the back half of the 2000s. In the UK as a whole, pay growth started slowing around 2003, but nobody really noticed that Scotland was bucking that trend. Its labour market had a really strong phase through 2005, 2006 and 2007, and it went into the crisis with higher employment and with typical pay having more or less caught up with the UK average. During the crisis, the effect on employment was slightly worse, but the slightly shallower and less extreme pay squeeze, particularly for the private sector, meant that, in general, the gap narrowed further. Scotland did what you might call well during what was a bad phase for everybody.

As we have discussed, we have more recently had a worse phase on the aggregate level. The question, then, is: what is going on in different phases to give that, and how do we square the circle of the fact that, although the numbers show record employment, people are pissed off? That is the less polite version of Patrick Harvie’s question. Why do we keep hearing about insecure work and low pay when everyone is saying that the aggregate data shows that everything is going really well?

The answer is more nuanced than what either side says. For example, people should be careful about saying that all this extra employment is bad, low-quality jobs. The data does not back that up, and it is really patronising to the people who are doing those jobs. If we consider who has benefited most from the increase in employment—which has increased significantly more than any of us thought possible—we see that it has disproportionately gone to lower-income households. That was not always true; in the late 1990s, the increase in employment went to second earners in higher-income households. That has not happened this time. It is progressive employment growth, and it has reached women, the disabled and the low-qualified. That is what you want to happen in Scotland, where the level of disabled employment is not too high, so you should be careful about saying that this employment is all bad.

I appreciate the attempt to paint a mixed picture, but on that last point, although one might argue that there are some benefits to the economy from such a change, surely it is not a good thing for individuals to have to work more hours to not even stay still in relation to inflation.

But what is the counterfactual? Would we all like to work fewer hours? History tells us that humans’ answer to that is yes. As we get more productive, we would like to work marginally shorter hours—although not quite as short as Keynes thought.

However, that is not the counterfactual. The counterfactual question is this: if hourly pay growth slows, is it a good thing that a flexible labour market allows us some scope to change that preference so that we can protect our incomes? It means that people are better off than they would otherwise be if they had to reduce their hours.

It allows some people that scope, but clearly it does not allow all people that scope. The fact that we still have high levels of in-work poverty at the same time as high levels of employment would suggest that the argument that work is the route out of poverty is broken.

I totally agree.

I thank Torsten Bell for giving us that very helpful picture.

“Flexible working” is a euphemism for people working at strange parts of the day—such as evenings or weekends—for which, historically, they might have been paid enhanced hourly rates. Instead, even in what might previously have been considered “good” public sector jobs, they now receive basic minimum hourly rates, which never used to be the case. For example, there is now not a single local authority in Scotland in which people are paid double time for working overtime on Sundays. Even for those working for mainstream employers, conditions of service beyond the hourly rate have deteriorated rapidly in the past 10 years or so.

There are more families in work who are also in poverty. Seven out of 10 children who live in poverty are in households in which at least one parent works. People are working flexibly, but for more hours because they are desperate to balance their domestic incomes—and employers in various sectors will take advantage of that. For example, in the social care sector, there is the sleepover issue, which the committee has probably covered before. In certain places, that does not offer even the minimum wage, let alone the living wage. People are doing part-time work not because there is quality work that they are choosing to do for lifestyle reasons; their reasons are economic ones.

People’s ability to move into quality jobs through upskilling or education and training, or into higher levels of jobs—for example, moving up through the nursing or social work professions—or out of menial jobs and into middle-level ones relies on staff training, employers skilling up their staff and the college sector delivering part-time and block-release courses. In the austerity period, the first budgets to go have been those for training. Very few employers are investing in their staff but, at the same time, they are saying that, for demographic reasons, they need to grow their own, and so on.

I am Unison’s lead for the further education sector. We have been pursuing a dialogue about introducing training schemes in the college sector at national level and, in the past, I have had discussions with Angela Constance on that. However, even the sector that delivers skills in education is not very good at skilling up its own staff. For example, if someone in that setting wants to go from being a janitor to an information and communications technology technician, they will find that very difficult.

I will add a little bit of nuance to what has already been said. The idea of people wanting to work more hours than they are currently being offered is the notion of underemployment. I have written quite a lot about that with my colleague Danny Blanchflower. It is certainly the case that, although unemployment is at a pretty historic low—lower than it has been since the 1970s—there is still more underemployment than there was prior to the recession.

We have also done work on the wellbeing of people in different employment states. It is certainly true that the level of wellbeing of those who are underemployed is worse than that of those who are fully employed—but then their level is better than that of those who are unemployed. Therefore there is a trade-off here as to which is the more desirable situation.

I turn to John Gallacher’s point about rates of pay. With a different colleague, I have been tracing the decline in paid overtime working since the beginning of this century. We have found that the use of paid overtime has declined pretty sharply. Of course, that also interacts with the minimum wage. People might be offered the minimum wage, but the counterbalance to that is that they are also offered a lower premium or fewer overtime hours. People being asked to work what we might consider to be non-standard hours, but for no more than the basic rate, is one of the big changes that have occurred.

When we talk directly to workers in different workplaces and sectors and look at the concerns that they raise, it is striking how consistently mental health, stress and job intensity are raised as an issue. The demand from union representatives for mental health first aid courses and suicide prevention services is very high. From the point of view of someone who spends a lot of time trying to service the needs of workers across the economy, I can say that we are having to look at such issues more and more. Occupational health services are having to do the same. An element of that trend can become lost, because it does not show up at all in pay figures, or very easily in questions about how people find work, but it is a scarring feature of the workplace across Scotland.

When we are thinking about how we raise productivity, we need to bear in mind that we have a workforce that is overstretched, who feel that they are not supported by their employer and are not being offered training and that they are running on empty the whole time. In those circumstances, how can we expect to bring about productivity gains? Improving productivity is a question of job design and supporting the workforce effectively, but when the levels of stress and mental health crisis among the workforce are so high, that is simply not tenable.

I go back to the point that, because women make up the majority of low-paid workers, they have greater experience of insecure temporary and zero-hours contracts. To a disproportionate extent, they are more likely to be on such contracts. Patrick Harvie asked about part-time work, which must also be considered. Women make up the majority of part-time workers, but part-time work is predominantly found in low-paid sectors.

The fact that there is often a cultural presumption across many organisations and sectors that part-time working is not suitable for higher-paid jobs sustains women’s concentration in lower-paid work. In addition to the problem of underemployment, which David Bell mentioned, many women are working below their skill level. Essentially, they are in the wrong job and could be working at a much higher level.

Helen Martin and John Gallacher mentioned the lack of training. In our work, women tell us that they have challenges in accessing in-work training, which prevents them from progressing into higher-paid work. On the whole, women are less likely to be able to access training, and women in low-paid part-time work are the group of workers in the labour market who are least likely to access training.

I come back to a point that John Gallacher made about influencing low-paid sectors. As members will know, it is now Scottish Government policy for sleepover hours to be paid at the same rate as daytime hours. Why would anyone disagree with that? However, the majority of the work that organisations such as Cornerstone do is commissioned by local authorities and partnerships, which do not have the resources to pay for people to be asleep at night if they have to be paid at the same rate as they are paid during the day. It is fantastic that there is ring-fenced funding for the Scottish living wage, because that has meant that our front-line staff have received a boost in pay, but all the local authorities are now taking away all the sleepover shifts. Our staff used to top up their income by doing three or four sleepovers a month. That has been taken away, so the increase in their salary from the Scottish living wage has been cancelled out, with the result that they have ended up in the same position.

That is what can happen if a measure is not thought through properly and there is not enough consultation with the sector about potential implications. Something that is very well meaning can have the opposite effect.

When employment rates are at record levels, we would usually expect that to create pressure for wages to be increased. All economic theory and past experience suggest that that would happen, but it is not happening. We have covered productivity, which is one aspect that is holding back that usual relationship, but there might be other ones. An obvious example is unseen slack in the economy. Although employment might appear to be very high, there are people who are underemployed or in insecure work. In addition, economically inactive people who are not counted as employed or unemployed are beginning to come back into the labour market, which is keeping wages down.

Self-employment is a factor that we ought to touch on. We have close to record levels of self-employment in the economy. It is very difficult to get hold of data on the earnings of self-employed people but, from what we can see across the UK at least, there has been a huge reduction in the wages of self-employed people since the crash. The profits of the self-employed have decreased by about 25 per cent. That means that the people who were already self-employed have experienced a reduction in wages or the new entrants are at a very low level. An increasing proportion of the self-employed are women.

The picture that we are seeing could be explained partly by the fact that we are counting people as employed when they are in low-quality work. As Torsten Bell said, there are pockets of low-quality, insecure work out there.

Let me pick up directly on that point. We need to be careful about saying that a tight labour market is not feeding through to pay growth. That is not true. It is not what the econometric evidence shows. It feeds through; it is just that the process cannot get going fast, because there is not a lot of productivity growth to feed through to pay growth.

It is not true that a tight labour market and some productivity growth will not lead to pay growth. That is not what the evidence shows. The evidence shows that we do not have productivity growth, so we are not getting pay growth. That is important to acknowledge, because otherwise we end up saying things like, “There’s no point getting productivity growth in the first place, because it won’t feed through.” We must be careful about what the evidence shows.

I want to go back to Patrick Harvie’s initial question. I agree with Anna Ritchie Allan. Why are we talking about all these low-paid, part-time, short-hours jobs in a way in which we should have been but were not doing in the 2000s? The answer is that lots more men are doing those jobs, so everyone has suddenly decided that we should talk about that all the time.

The big picture on low-paid, low-qualifications work is that women have done and still do a hugely disproportionate volume of it. However, the increase is all about men and the decrease is about women. The levels are high for women, but the change is that men are going into such roles and women are coming out of them as there is general occupational upgrading for women over time. That is probably a large part of why we started talking about the issue, for the reasons that Anna Ritchie Allan gave to do with structural issues, people’s attitudes to work and other things.

The issue is driving quite a lot of the wider changes that we are seeing in the aggregate wage data. What is pushing up earnings inequality, which is one of the issues at the moment, is not that the highest-paid people in Scotland are getting particularly high wage rises but that lower-paid men are seeing hours reductions, relative to where they were 10 to 15 years ago. Higher-paid men are not seeing those reductions.

We are therefore seeing increasing inequality in hours. Traditionally, lower-paid, working-class men did the longest hours in the economy. They were supporting a family by doing 50 hours or so and being paid overtime for doing that. That was the structure of the economy, but we have been moving away from that for some time.

It is those people who are really angry. Our qualitative work shows who is least happy with the world of work, as it exists. Even though women are doing most of the low-pay, low-valued—wrongly low-valued—work, men who are doing such work did not see their dads doing such work and did not expect to be doing it.

The change is very large. The share of low-paid work that is being done by men is up about 45 per cent since the turn of the century, so we are talking about big changes—and they are big social changes; they are not just to do with the economy.

The question is how we wrestle with that. It is hard, because better sharing of low-paid work between the genders is a good thing. We want less low-paid work overall, but when we are measuring pay relative to the median—as something that is not going down and is just an average, to some degree—we want such work to be gender shared. The counterpoint to that is that the situation is causing a lot of discontent. That is where these things are hard.

We will probably come on to the gender pay gap. The process that I talked about needs to continue, but the issue feeds into Anna Ritchie Allan’s point about where the on-going, large gender pay gaps come in, which is basically when people have kids, are forced into working part time and then stay in part-time work because of the lack of flexibility in higher-paid roles. That is the dominant factor, to which we have no adequate policy response, as yet.

As a general point, we need to pay more attention to hours worked—who is working what hours—if we are to understand what is happening to our country. It is really worth doing that. People look only at the income levels, because they are thinking, “Everyone in this world gets a salary, right?” rather than thinking about hours worked. Understanding hours worked is crucial to understanding how income distributions change over time.

Russell Gunson’s point about self-employment is an important one. It is important for Scotland’s tax revenues, because we are now up to around 14 per cent of the workforce being self-employed. The growth in that has largely been among self-employed people who do not employ anybody else and just work on their own account. Typically, they earn very little, so the spread of the income distribution of the self-employed is wider than the spread of the income distribution of the employed. Some self-employed people do very well, but there are many who do pretty badly. They have no form of representation, because they work on their own behalf.

Some of the work that we have been doing recently—without coming to any conclusion—is to try to understand why self-employment has been growing so fast in the UK, whereas in the States the number of self-employed is falling. Those are two countries that apparently have high employment and low unemployment but have completely different trends for self-employment. What is it about the UK that makes self-employment seem like an attractive option?

On an unrelated point about training that comes up quite often, some of the discussion in the UK about the lack of productivity growth has been about management quality. Scotland’s management quality, when compared to that of our competitor countries, needs to be given some attention. One of the ways we see that is in a lack of training for employees.

I thank everyone for their input so far.

Torsten Bell, you mentioned the depreciation in the value of the pound. I saw the statistic that the pound’s value has reduced by 18 per cent since the EU referendum on 23 June 2016. I am interested in the care and agricultural sectors, which Helen Martin mentioned, too. We are seeing that people are not coming from the EU to work, especially in the dairy sector—48 per cent of Scotland’s dairy farms are in the south-west. Is there evidence that people are not coming to work in the dairy sector or in the care sector because the pound is not as valuable as it was? Is evidence emerging that agricultural workers are not coming here and are instead going to France or Germany where it is euro for euro and the pay is better?

I cannot give you a definitive answer to that because the data does not provide it, but you will all have heard anecdotal evidence of employers saying, “It is a disaster; I can’t hire anyone at the moment.”

What can we see in the migration data? There has been a change in behaviour since the referendum. There have been decreases in the number of people coming here from the accession countries and increases in the number of people leaving who come from older bits of the EU in western Europe—the original EU15.

What drives that is more complicated. The pound is definitely part of it; if people are thinking about coming here temporarily to do a job for a year, that is now less worth doing. However, do not underestimate the fact that wages in Poland have been growing fast over the past five years—that is a big part of what is going on. The worst of the euro crisis is over, so the alternative labour markets that are out there have changed fundamentally. Plus, we have sent out a message that we are less keen than we were, so it is not massively surprising that, on a human level, people have taken different lifestyle decisions. You will all know people who have made versions of those decisions.

Is there good evidence that distinguishes between those different effects and says that this bit is because of a stronger economy in some parts of Europe, this bit is because of the depreciation in the value of the pound and this bit is because of attitudinal issues? No. I warn slightly against saying that it is all about depreciation, because the depreciation in 2009 was bigger than the depreciation after Brexit, but there was a surge in migration from the EU after the 2009 depreciation and we now see the opposite. The depreciation is definitely part of it, particularly for people who are short-term migrants and come here in order to send money back, rather than because they are making a bigger choice about their life. However, we cannot say that it is the whole part.

Also, just as a general word of warning, the migration debate is now seen totally through the lens of Brexit so, if you are pro-Brexit, you say that the migrants were causing all the problems beforehand, and if you are anti-Brexit, you say that any fall in migration is a disaster and the economy will immediately go off a cliff. Neither of those positions is backed up by the economic evidence. They are both politicised positions. People may hold those positions for other reasons—they may want migration for what it does to our society, creating a more open feel, or they may want the opposite—but they are not based on economics.

Employers will always say, “I can’t get the staff I need” but the first thing I always say to them is, “Have you increased the wages? Have you tried the bold approach of offering a pay rise to see whether anyone comes to work for you then?” The answer is always, “Oh, no, no.” Employers do not have an impartial view on migration policy. They have a slight vested interest in the situation. That is all that I am saying.

The numbers are down for some combination of those three reasons. I have no idea what the balance between the three is.

For us as an organisation, we have not noticed any difference, but that is because of the type of organisation that we are. We provide a lot of support in the community and in people’s own homes. I do not have access to the data but if you speak to Scottish Care, which represents a lot of the care home providers, I suspect that it will have seen a change in the number of the EU nationals coming over. I do not have that data, but it would probably be able to give you that data.

Very briefly, because Torsten Bell has anticipated what I was going to say a little bit, the question is more about what happens next. We would argue very strongly that this is an opportunity to look at the low-wage sectors that have relied too heavily on migrant labour and start looking at job quality, job design and pay. We would put it back to the employer quite strongly, as Torsten just did, and say, “If you can’t attract skilled workers, you have to ask yourself a serious question about why that is.” We simply do not accept an answer that says that the only way to run a business is to run it on very low wages and migration.

Okay. We move on to our last theme, which is distribution of earnings and implications for revenue.

We have covered some of this ground already, but I want to focus on the distribution of earnings and the implications for revenues that accrue to the Scottish Government, which principally means income tax rather than national insurance. I will focus mainly on employment rather than on self-employment, because of the current distribution of devolved taxation.

I want to look at both high-paid and low-paid jobs. Perhaps we can start with low pay. As the committee reported in its most recent report on the Scottish Government’s budget, in a country of about 5.4 million people, there are 2 million people in Scotland who pay no income tax at all. Is that desirable and is it sustainable? If it is neither desirable nor sustainable, what should be done about it?

We have quite a strange structure, because you cannot really think of income tax separately from national insurance. National insurance kicks in earlier than income tax does, at somewhere over £10,000 or so. There has always been an argument for amalgamating national insurance and income tax, although civil servants suggest that it would be very difficult to do that. That has now been rendered more difficult by the devolution settlement. When national insurance and income tax rates are combined, we end up with a profile that looks quite strange—it certainly does not look smooth. In fact, the lack of change to the higher-rate allowance has produced an unusual spike in income tax and national insurance recently.

It seems to me that there is an argument that by not combining those two taxes, we end up in a situation of great difficulty, which will cause tensions as far as the devolution settlement is concerned. Whether we can move to Scotland having powers over national insurance remains to be decided by politicians. Compared with other countries, we have a very high personal allowance, which the Scottish Government has no control over. We do not know at the moment whether that affects work incentives, but it is worth further consideration. That would have to be done in conjunction with national insurance, because I do not think that people make the distinction in their pay packet—it is not clear to them.

You asked about sustainability and desirability. The issue of desirability comes down to what you are attempting to achieve. For example, if you are trying to reduce child poverty to the levels that everyone in this room has pledged to do by 2030, an increase in income tax at the bottom end may counter that. As the universal credit taper impacts a great many people at the lower end, would it be desirable to tax on top of that taper? The rate could reach 75p in the pound; if we think that 46p in the pound causes problems for work incentives at the top end, surely 75p in the pound would do that at the bottom end?

Therefore, desirability comes down to a judgment call on what you hope to achieve. The personal allowance is a function, in essence, of a UK-wide decision to remove many people out of income tax, supposedly—if that word is not too judgmental—to be progressive, when in fact it benefits nobody under the personal allowance threshold. It is a big and costly spend—£110 billion or £120 billion across the UK—most of which does not go to the very poorest.

The sustainability question is interesting. As Torsten Bell has said, the system across the whole economy and tax system is not particularly progressive. We have a very progressive income tax system, in technical terms, and in Scotland that is leading to a big dependence on a very small number of higher-rate and additional-rate earners. No doubt we will come on to that.

We will get on to the higher end, but one solution at the low end would be to focus on people who are in work on a minimum wage to push them much closer to the median wage. That would have implications for our skills system, such as focusing on those who are in low-paid and insecure work, and for productivity and the economic strategy, which we have touched on, with a focus on the parts of the economy that pay the minimum wage to try to raise people’s pay.

Therefore, the answer to the sustainability question is more cut and dried; some of the responses are about boosting pay—for the low end at least—which, in turn, would deliver the inclusive growth that we all suggest we wish to see.

There is still a lot of in-work poverty. We work with the Child Poverty Action Group to produce reports such as the one that I have with me, on “Fair Work and Decent Childhoods”. A short anecdote about a recent development does not answer your question, but I mention that none of the local authorities has consolidated the living wage into their pay structures, so £9 an hour is not the substantive least that a person in a local authority can earn.

A lot of local authority workers have received the recent 3.5 per cent pay deal, which is not an astronomical amount of money. When those for whom universal credit was in place got their back pay to 1 April 2018, they lost their benefits and were put into reassessment. That meant that their actual household incomes reduced despite the 3.5 per cent increase, because the Department for Work and Pensions counted the back pay as earnings in that one month. Something is going wrong in the balance between tax, benefits and salary. At that level, a pay rise has punished those people with regard to their household incomes.

Constituents have contacted me specifically about that.

As Unison represents workers in the public sector, I presume that John Gallacher was talking about a majority of women workers in that group. It is hard to overstate the impact on women’s incomes of social security and so-called welfare reform. A reason for women being so overrepresented in lower-paid jobs is that greater reliance on social security, because they are in low-paid jobs and they shoulder the burden of unpaid care for adults and children. It is critical to take a gendered approach in considering income tax and the position of women in the labour market.

Our view is that it is not desirable for the personal allowance to be so high. It has cost about £10 billion, so it has been very expensive to achieve, and the increases to the personal allowance—particularly the most recent ones—have disproportionately benefited higher earners. There are other arguments for that position that the Resolution Foundation as an institution would not take.

I will comment on the issue of what you believe about a stakeholder society and how you want that to feel. Over time, income tax historically—that is, pre-war—was not paid by a majority of the population. Instead, inheritance tax was paid by the majority. Hilariously, we have gone to the opposite extremes and started to come back the way on income tax. Societies take different decisions at different times. We should be really clear that we are taxing the income of people, even if they are not paying income tax.

Scottish politicians have ended up with a certain tax base. Overall, is it desirable to have an income tax base that is not certainly but potentially more volatile? All else being equal, that is undesirable from a fiscal perspective, although you could compensate for that in other aspects of the tax system. However, taken in isolation, I think that that is undesirable. The tax base might not always be more volatile. You could imagine a system in which low earners suddenly have very volatile earnings, but that is less likely to happen.

If your tax base is disproportionately made up of that bit of the tax system, clearly you have ended up, as a by-product of the personal allowance policy, with a more volatile system for the tax base that supports Scottish Government spending. All else being equal, that is undesirable. Whatever you do about that will be hard to do. I support everything that others have said about bringing people up nearer to the median for earnings and stuff, but doing that—I do not want to be too blunt about it—will not get you very much income tax revenue. I will give you an example. Possibly 40 per cent—it might even be up to 50 per cent now—of the household income distribution does not pay much income tax at all. All the cuts to income tax are of no use to them, and they are being hammered by the big cuts to benefits, which is why child poverty will rise over the next few years.

Women will be losing from the work incentive effects of universal credit. That is not the big thing that is driving down income to the bottom third of distribution; the big thing doing that is the benefit freeze and the cuts to support for large families. Universal credit is not the problem—it is the other cuts to the social security system that are reducing incomes. There is nothing that you can do on the tax system to compensate for that, because the effects are so large.

If I ignore housing costs and refer to people’s disposable income after they have paid their rent, once the welfare cuts are rolled out fully some families at the bottom will be losing 15 per cent of their disposable income from the cuts. That is a massive impact, so tax tweaks for most of those families are neither here nor there.

There are some work incentive effects, but I would not want to go the other way. Some people argue for there to be no personal allowance. If they take the extreme version of my argument, which is that increasing the allowance as much as it has been increased is regressive, they might say that we should not have a personal allowance, and that we should scrap it and replace it with, for example, a universal basic income. I caution against having no personal allowance because, with regard to the incentive to enter work, you do not want income tax to kick in as soon as marginal workers enter work. That would have a big effect on the incentive to work, particularly for women, that I would be nervous about.

As I said, the level of personal allowance is now too high, but I would not want it to be zero.

That is fascinating. Obviously, we are focused on revenues, but the issues that Torsten Bell has raised about stakeholder society are important to us, too.

We hear quite a lot about inclusive growth, but we do not hear so much about inclusive taxation. I wonder whether that is something that we should think about.

I want to turn the discussion around and think about the top end of the labour market. We have heard a lot this morning, and in your written evidence, too, about wage stagnation, and how there have not been very significant wage increases over the past 10 years or more in any part of the UK, including Scotland.

So much of what the committee does is analysing forecasts that are given to us by the Scottish Fiscal Commission and others. It is very important to the way in which the fiscal framework operates that we understand where taxation is likely to go. Notwithstanding the wage stagnation that we have seen over the past 10 years, the number of additional-rate taxpayers and the number of higher-rate taxpayers are forecast to grow very quickly over the next five years. The number of additional-rate taxpayers—I am talking only about income tax—is forecast to increase by more than 30 per cent and the number of higher-rate taxpayers is forecast to increase by nearly 25 per cent between now and 2023, notwithstanding the fact that we have seen no real wage rises over the past decade.

I have two questions about that. First, what is driving those forecasts? Secondly, they are just forecasts, but in your experience and your judgment, how reliable are they likely to be?

The figures are appearing in those forecasts because we do not uprate the thresholds. That is what is driving your additional rate—the £150,000 is set, as it were, and does not rise with inflation. Even if earnings overall are growing by only 1 per cent above inflation, if you are not even uprating the threshold by inflation, clearly more people are brought into the net. That is what has driven the increases. That is for the additional rate, but be careful because those increased percentages are based on 10 people going up to 13—there are not lots and lots of people in the bucket.

It is like £15,000 going up to £20,000.

It is a big percentage number, but the absolute numbers are not that large—although it matters for tax policy, because they pay a lot of tax.

On the higher-rate group, which is a much larger percentage of the population, the increase is being driven by a policy from pre-2015, which is when the personal allowances rises happened. The higher-rate threshold was sometimes cut intentionally to avoid the personal allowance benefit feeding through to higher-rate taxpayers. That was stopped around 2015—we moved to increasing the higher-rate threshold. That is now not true in Scotland, but it is true in the UK.

You have a choice. If you want revenue to come in and you do not want to change your tax rates, there is fiscal drag either in the economist sense, which is earnings being faster than inflation, or in the brutal politics sense, which is in absolute terms freezing your thresholds and then allowing even inflation plus earnings growth on top of that to drag more people into higher rates over time. You have straight choices about where you want the revenue to come from.

In some ways, from an economist perspective, we do not want lots of very high marginal tax rates in and of themselves, but there are very real trade-offs about where the revenue comes from if it does not come from those sources. I am nervous when people say that there is some kind of absolute limit to how many people can pay any particular given rate. Sensitivity to tax rates is very different for different kinds of people. The behaviour of single parents and second earners in couples is very sensitive to tax rates and I would be very careful about how universal credit interacts with tax rates at the bottom for those people. You might think that this says it all, but high-paid men are insensitive to tax rates in their decisions and will basically do what they were going to do anyway, more or less. What you care about depends on who you care about.

I just want to back up Torsten Bell. The freezing across the UK of the additional-rate threshold, even though that is devolved, will drag more in. The Scottish Government’s policy of freezing, or at least having under-inflation increases in, the higher rate will also be a big factor in broadening the tax base at the higher end.

We supported the freezing of the higher-rate tax threshold in Scotland for this and future years for two reasons. One of those reasons was about increasing tax revenue and, to return to the points that were made at the start of this session, being able to invest further in public services and anti-poverty measures. The other was to broaden the tax base so that we are reliant on a greater number of people at the top end, too.

The point about fiscal drag is well made. Clearly, if the thresholds are held constant, earnings growth will take quite a few more people into the higher tax bands. It is important for the Scottish Government to think through the interaction of national insurance and income tax. The strange 52 per cent rate is being paid by those on upper but not really high incomes.

On Adam Tomkins’s original question, I am struggling to think of businesses that will generate extra high earners. There will not be extra additional-rate taxpayers in the public sector. I assume that they will be entrepreneurs with new, fast-growing businesses. It is admittedly hearsay, but I do not see huge evidence of that.

A number of us are sceptical about the forecasts, especially when the Scottish Fiscal Commission forecasts that earnings growth in Scotland will be lower than in the rest of the UK and the employment rate will also be lower. Those are just forecasts, but matching up the various forecasts proved a challenge when we looked at the budget earlier in the year.

I thank the participants for what has been a fascinating and informative session. There have been great contributions from all around the table. The session will certainly help the committee to have a good grounding in what we should ask the Scottish Government when we hear from the cabinet secretary on 8 May. Obviously, much of this is outwith the powers of the Scottish Government. The UK Government is also involved, and that might be something for the committee to think about for the future.