HoC 85mm(Green).tif


Treasury Committee 

Oral evidence: An Equal Recovery, HC 152

Wednesday 15 September 2021

Ordered by the House of Commons to be published on 15 September 2021.

Watch the meeting 

Members present: Mel Stride (Chair); Rushanara Ali; Harriett Baldwin; Anthony Browne; Dame Angela Eagle; Julie Marson; Alison Thewliss.

Questions 1 - 74


I: Torsten Bell, Chief Executive, Resolution Foundation; Liz McKeown, Director, Public Policy Analysis, Office for National Statistics; Christopher Snowdon, Head of Lifestyle Economics, Institute of Economic Affairs.

Written evidence from witnesses:

Resolution Foundation

Office for National Statistics


Examination of Witnesses

Witnesses: Torsten Bell, Liz McKeown and Christopher Snowdon.

Q1                Chair: Good afternoon and welcome to the Treasury Select Committee and our witness session as part of our equal recovery inquiry. We are expecting—in fact, we are absolutely certain—that there will be a Division of the House at around 4 pm, whereupon I will suspend the Committee, but don’t go anywhere, witnesses, because we will very much want to have you back after that. I will suspend the Committee probably for about 15 minutes in total.

I am very pleased to be joined by our panel of three witnesses. I wonder if I could start by asking each of you to very briefly introduce yourself to the Committee.

Torsten Bell: I am Torsten Bell. I am the chief executive of the Resolution Foundation.

Liz McKeown: I am Liz McKeown. I am the director of public policy analysis at the Office for National Statistics.

Christopher Snowdon: I am Christopher Snowdon. I am head of lifestyle economics at the Institute of Economic Affairs.

Q2                Chair: Welcome and thank you all very much indeed for giving us your time this afternoon. I want to focus on income inequality to start with. It is really a question for all of you. To what extent has this pandemic exacerbated existing inequalities in the UK, and how lasting do you think these impacts are going to be?

Torsten Bell: It is a big question. At one level, the answer is dull, which is that the income effect of the pandemic is not a huge change in income inequality. It certainly is not a large increase in income inequality, which I know is what some people feared and expected would happen. That is despite lower earners being much harder hit. Because of the spread of earners across the income distribution and the fact that the bottom of the income distribution has fewer workers in it, the poorest parts of the population have fewer earners per household—

Q3                Chair: Why did it not happen, when it was expected?

Torsten Bell: I will go through the reasons. Step back: what is the nature of the crisis? It is a labour market crisis and the poorer parts of the population are less exposed to hits to their labour market income than middle and higher-income households. Even though lower earners are hugely disproportionately hit, the feed-through of that labour market shock to the household income distribution is much more even than you would expect, looking at the headlines for the earners. That is the first thing.

Secondly, the Government policy response did a lot to reduce the inequality of the household income, and particularly boosted the incomes of poorer households via benefit increases, and then in general squashed the scale of the feed-through from labour market shock into household income shocks via the furlough scheme and other mechanisms. Those are the two big reasons. I have to say, if you just looked at the household income distribution of working households, so taking out non-working households, you would get a slightly higher preponderance of damaged incomes for lowerincome households.

The most useful way to look at it, in some ways, is that the big aggregate impact of the shock is less the big changes to household incomes, given that the Government’s policy response, on average—really importantly on average, not in each individual case—protected people. What actually happened were changes on the consumption side, and that was very unequal. The nature of the shock is felt on the income side as a labour market shock. Obviously its driver is changes on the consumption side, in that we made it illegal to spend in some ways and we did not want to spend in those ways anyway.

That fed through into big changes in consumption patterns. Those are very uneven across the income distribution. At the top, they mean big falls in how much people are spending, and that is particularly true if you look at foreign travel. The amount rich families in Britain spend on foreign travel is very large, and that stopped overnight and they could not spend it domestically. There are only so many DIY things they could buy for the house, so savings went through the roof and the gradient is very steep.

If you look at lower-income households, in lots of cases, particularly if they had kids, they saw their spending actually have to increase, because of suddenly having to have the kids at home. Lots of the ways in which lowerincome households cope with periods of financial distress, be it going to the library for free ways to keep your kids entertained, or going to your mum’s when you need to feed the kids and you have run out of money that week, were not available during this crisis. If you look at shopping patterns, it was harder to go to cheaper shops in some bits of this crisis.

So lower-income households did see increases in spending, which we did not see for the rest of the distribution. That is why you are seeing high savings at the top—much more likely to save than lower-income households—and increases in debt at the bottom end of the households. What I am basically saying is that income inequality did happen, but for the reasons I have outlined is flatter than you might expect across the income distribution. Do not look to the income inequality measures for the main inequality of the crisis. Look at the spending side, and that then feeds through into household wealth, because you saw higher savings at the top and higher debt at the bottom. That is the lasting inequality effect of the crisis.

Q4                Chair: Can you quantify in any sense the difference between the top end and the bottom end on the savings?

Torsten Bell: Yes, we have done that. The ONS has also produced some great work for early sight for some of its household surveys. I am happy to send that across to the Committee. We have set out the gradient in a number of surveys over the course of the last year. The gradient is steep. The really big difference is in the amount. You are more than twice as likely to have increased your savings if you are at the top than if you are at the bottom, for example, and obviously you had more savings. You are increasing savings there in a very large number.

When the Bank of England come to talk to you and talk about the aggregate savings wall that is sitting there, which we are all relying on to drive consumption for the coming years, that is savings held by people who do jobs more like the people in this room and less like the people who do jobs in retail, hospitality and leisure.

Q5                Chair: Would it be the case that, if savings have gone up at the top end of the distribution, and those individuals would have a lower propensity to spend those savings now than perhaps those at the bottom, that in turn has had an impact on the consumption support that this pent-up saving has produced? Has that worked its way through in any sense? Have you looked at that?

Torsten Bell: As a hypothesis, it is almost impossible for it not to have an effect, because different-wealth families and different-income families spend their money in very different ways, in terms of what they purchase. From the limited consumption data we have so far, I do not think you can show that conclusively. As I say, to do that well we would need to wait for some ONS survey data that I fear we are not going to see for 18 months, probably more. We will need to get into this year, when they can actually spend money, so it may be two years away. To show that conclusively is going to be hard in the live time, so no is the answer, I fear.

Chair: Can I turn to you, Liz, for your observations?

Liz McKeown: Torsten is right. The definitive picture on some of these issues is to come. We will see in January, with our core income inequality publication then, some of the major impacts here. It has been important for us in ONS to draw some earlier insights from the whole range of surveys we have. During the pandemic, we have been really keen to get some of those earlier insights on these sorts of issues out and launch new surveys that allowed us to get more timely information than we have had in the past.

To pick up on some of the points that we have already begun discussing, what do we know about what has been happening to income from some of those earlier findings? We know that workers on lower incomes are more likely to report decreases in income in the year to March 2021 compared to the previous financial year. To give you a scale of those differences, that was 42% of those on the lowest income reporting a decrease, compared to 31% for those on the highest income. We are already seeing that come through some of the early figures.

We know that those on lower incomes were more likely to be furloughed. They were less likely to be able to work from home than those on higher incomes. We also know that people on higher incomes were more likely to be paid in full if they were unable to work, compared with those on lower incomes. Across the income area, we are beginning to see those differences play through.

Torsten is right: we also see these differences in spending. It is important to look at both income and spending. There, in the figures, we are seeing some of the trends that have been pointed out. Higher-income households saw the biggest reductions in spending during the pandemic. Households in the bottom fifth of the income distribution saw a reduction of 12.5%, compared to those in the top fifth, which saw a reduction of 20%. Clearly, in absolute terms, for the people in the top fifth it was still an absolute larger reduction, because their income was larger to start with, but those percentages are different in that way.

We know that those households on lower incomes are typically less able to cut back on discretionary spending. The international travel is one example. Households in the lowest income decile spent 54% of their total weekly expenditure on essentials. That is things like housing, food and transport. That 54% compares to 42% in the highest income decile. You are seeing these differences in spending.

We also know that, because higher-income households were more likely to shift to homeworking, that also would have had an impact on their spending. If you take those changes together, the income and the spending changes, you are seeing that those on the higher incomes had greater opportunity to save and to ease any financial pressures that there were. Although we do not have a definitive picture coming from the statistics at the moment, there are a lot of early indications of the sorts of changes we are seeing across the income and spending space.

Christopher Snowdon: Being forced to save money is not necessarily a benefit. Presumably these people would rather have spent the money in the first place. The fact that they have had to defer that spending does not mean that they have disproportionately benefited. You could say they have been disproportionately harmed.

On the question of whether inequality will rise, I would echo what has just been said. We do not really know yet. It is not over yet. Apart from anything else, a lot depends on unemployment. If we look at previous recessions, we see that inequality tends to either stay the same or decline. That is generally because benefits, at least since the early 1980s, have been fixed to inflation rather than to incomes.

However, this is not a normal recession. Clearly it is much deeper, less long lasting and we had the furlough scheme. I was very concerned about very high unemployment a year ago. I am much less concerned about that now, but it still has gone up. We should bear that in mind. I think unemployment was down to about 3.7% or 3.8% before the pandemic and the figure yesterday was 4.6%. It is still historically quite low but, nevertheless, that is quite a lot of people still out of work, and it is possible there will be more.

At the end of July, 1.6 million people were still furloughed. We have more than 1 million vacancies for the first time in history, it seems. Perhaps, if we can combine all these things, we should be okay, but it is not certain that we will. It is not certain that we can match up people’s skills and we seem to have a shortage of foreign labour. I am not quite sure to what extent that is Brexit and to what extent that is Covid and travel restrictions. It is probably a combination of both.

I am trying to say, I guess, that, when the furlough scheme unwinds, we will know where we are. If there is very significant unemployment, and it is at least theoretically possible, we could have another million or so people out of work. I do not think that will happen, but it is possible. We do not know whether the jobs that were put into deep freeze are still there. That is the bottom line.

Even if there turns out to be a surge in unemployment, it is by no means certain that that will lead to a rise in inequality. As I said, you have not seen that in previous recessions. You did not see that after 2008. You did not see that in the early 1990s or the early 1980s. Inequality is not a very good measure of economic wellbeing in that sense.

Q6                Chair: Thank you. That was very interesting. What do we think is going on with wages at the moment? We have seen increases in wages and some of that has been benchmarked against depressed figures from the past for the reasons we know. How is that going to play out across the income distribution? What do we think is going on there? Who is getting, on balance, the better end of increased real wages at the moment? What is the extent of that real wage growth likely to be going forward?

Torsten Bell: Since you asked about the facts on how much more people were saving, high-income individuals are four times more likely to have increased savings during the crisis, not twice. I want to correct that.

On earnings, the first thing to say is the data is rubbishnot just the ONS data, but the data is pretty rubbish at the moment. It is not just for the reason you are mentioning, which is the baseline problem. The baseline is a year ago and a year ago we were busy having negative earnings growth in lots of cases and a low level. It is also because of the compositional shift away from low earnersmore low earners leaving the labour market disproportionately. Although no one’s wages are therefore rising, the average wage rises.

The two effects together are explaining something like half to slightly over half of the measured wage growth we are seeing in the average weekly earnings figures. We, the Bank of England and the ONS have all had a go at correcting for those effects. They show you wage growth probably slightly stronger than we had overall expected a year ago, so something in the 3% to 4% ballpark, depending on what measure you want to use, slightly lower if you look over a two-year averaging approach, rather than trying to correct the data within it.

It is reasonable. I do not feel any confidence in giving you a point estimate of what the right level of what is happening to wages is, in order to compare it to a normal recession or a normal phase. Some real earnings growth is happening right now and that would not have been true in all recessions, so that is noticeable.

A large part of that comes down to Government support. If we step back and ask, “Why do wages fall in a recession?”, obviously firms not having the cash as demand falls is a large part of that. We have not seen many insolvencies during this crisis because the state has stepped in to support the cashflow of all kinds of firms. A socialisation of that risk has protected workers’ wages indirectly, either by the furlough scheme or by supporting firms’ cashflow. That is what we have definitely been seeing there.

You are then getting to this distributional question. How much is it true that we are now seeing some kind of new dawn for worker power because people can negotiate for higher wages because there is a huge shortage of workers? My annoying view on that is that it is true in some sectors, but it is overdone as an argument about the labour market, considered collectively. On the latter point, the danger is that we are ignoring the fact that we are still working considerably fewer hours than we were before.

There is the idea of this really tight labour market when, as Christopher says, we do not know exactly the increaseI would have thought something in the low hundreds of thousands, as people flow off furlough in two weeks’ time. I think we are currently at about 2% down in terms of hours worked. That is a bit less but broadly in line with what you might see in some recessions. There is not a small fall in hours worked; it is a lot. I am nervous to conclude, which some people have, that this is just some really tight labour market we are seeing.

There are some sectors that were migrant reliant to a significant degree, and ones that combined that with having more risk from face-to-face contact. People might not want to go and work in those sectors to the same degree, completely reasonably. I spent a lot of happy years working in kitchens, restaurants and things. I am definitely less keen doing it right now. That is not just because I am older. It is because there is a lot of Covid going around. In those sectors, you are seeing some responses.

We need to be careful. Not all low-waged workers are seeing huge pay rises. Cleaners’ wages are flat. The minimum wage is what is driving up their wages, not some kind of hugely tight labour market. I hear a lot of boosterism on the radio, with people saying, “It is boom time for all these workers”. Let us wait and see. In the US, there is a bit more evidence of that.

Liz McKeown: We are seeing the labour market continue to evolve. Yesterday’s figures showed the number of payrolled employees back to where they were before the pandemic started. One interesting thing is that in those groups that were disproportionately impacted at the beginning of the pandemic, in particular thinking about the young, those living in London and people working in hospitality, we are seeing the strongest growth now.

We have that on the one hand, the high level of vacancies on the other and the unfolding of furlough. The next few months of labour market data will be really key to see how those things interact and play out together. We will begin to get a much better picture of some of this then.

Christopher Snowdon: The labour market genuinely is quite tight and I would expect it to be quite tight for quite some timethe next six to 12 months, let us say. If most of the people who have been furloughed go back to work in the same job, I would think it would be quite a good time actually for workers to get a pay rise, particularly at the lower end, not universally by any means, as Torsten says. There are over a million vacancies and I am assuming that the migrant labour shortage is not going to end imminently.

I think we will see some decent wage growth, actually. The only question is how much of it gets swallowed up by inflation. If we get to 4% or 5% inflation by the end of the year, which is quite likely, we may see the same thing happen as in 2016-17, where the economy was doing quite well, incomes were starting to go up, wages were starting to go up, and then the whole thing got swallowed up by the pound falling and by inflation.

Torsten Bell: That triggers two important things. First, when looking at any of the labour market data at the moment, it is really important we do not treat it like a normal recession. The pandemic nature of what is going on in the labour market explains almost everything that looks very weird in the dataso switching half the economy off and turning it back on again. The matching period, to get people matching in the right places to the right jobs, takes time and we should expect that.

Secondly, this inflation point is really important. Worker power in Britain is generally low. When you get an inflation spike, which is what we saw after the financial crisis and after the Brexit referendum, it feeds through almost directly into lower real wages. That is what caused the wage squeeze of the last 10 years. Our response to that is that we all choose to want to work more to cope with the income shock. That is why women’s employment has gone through the roof since the financial crisis, largely. That is the main driver: big income shocks to family incomes from that inflation squeeze.

We did not see that during this crisis so far. We are about to see it. It is a really important question. We do not know how far, but our numbers are 4% by the end of the year, largely transitory, but if that transitory ends up being a longer time period then you will see this effect on real wages. If that happens, I do not think we should just assume wages will respond to that. If you think wages are going to respond to that, you are the kind of people who think the Bank of England should be hiking rates now, because that is how you get a wage-price spiral. In the real world, trade unions do not run Britain anymore.

Q7                Chair: I want a one-word answer, but I think I have had your one-word answer, on inflation, very quickly. To the other two panellists, yes or no, do you think the Bank of England’s assumption here of around 4% by the end of the year and then coming back down, in other words that it will be transitory rather than persistent, is correct?

Liz McKeown: It is not one I would respond to myself.

Christopher Snowdon: No, it is too optimistic. It has been too optimistic for a long time.

Chair: There is so much more I would have liked to ask you, but I have gone on for a while.

Q8                Dame Angela Eagle: In its evidence, the TUC has said that the pandemic is hitting low-paid workers disproportionately hard. Statistics show they are also more likely to have contracted Covid and died from Covid. Is that accurate?

Liz McKeown: As I think I said earlier, lower-income workers were more likely to be furloughed and more likely to report decreases in income.

Q9                Dame Angela Eagle: This is the 42% decrease in income that you referred to earlier, compared to the 31% for those on higher incomes.

Liz McKeown: Yes, exactly.

Q10            Dame Angela Eagle: Is there any work on the effect on mental health and wellbeing that that kind of precarious living has on a worker? For those who did not lose income and could work from home, it was inconvenient, but they managed to save, as you all pointed out in earlier evidence. They managed to save more because they could not spend it, but I imagine your experience of life is very much more difficult if you are a key worker or you work in the care sector and you have to keep going out, or if you have suddenly lost your job and you have no income.

Liz McKeown: No, absolutely. We picked up some of those impacts through our opinions and lifestyle survey, where we looked at the impacts of Covid on different groups. To give an example, we saw that those who experienced a loss of income were also likely to report financial pressures and difficulty making ends meet. We looked at wellbeing of different groups as well and saw how different groups were impacted during the pandemic differentially.

We can provide more detail. We have a great wealth of information on those wellbeing impacts across all different groups, from the demographic characteristics through to people who are on lower income. We can provide more detail on that, but we saw differential wellbeing impacts throughout the pandemic. It has been important to be able to capture not just the differential in the economic impacts of the pandemic. We also saw, as you said, differential health impacts and differential social impacts, in terms of wellbeing.

To consider another dimension, we looked at the environmental space. Things like access to green spaces varied as well.

Q11            Dame Angela Eagle: They were of course closed during the first lockdown. If you did not have your own garden

Liz McKeown: Yes, exactly, and that varied quite significantly. For example, different ethnic groups had vastly different access to their own private green spaces. We see that coming through in the statistics as well. Across the different dimensions you might want to look at when thinking about both the pandemic and the recovery from it, you are seeing those differential impacts across different groups.

Q12            Dame Angela Eagle: It would be good to have more information on that. Torsten, has the financial insecurity of low-paid workers increased since the pandemic? Is it likely to increase in the future, or is the pandemic just one of those shocks? To some extent, Brexit going on underneath is reshaping things. Do you think that financial insecurity for low-paid workers has increased? Is this something that policymakers should be looking at?

Torsten Bell: The crisis was sectorally focused. That is the thing that is unusual about this crisis compared to other crises.

Dame Angela Eagle: So it is hospitality and leisure.

Torsten Bell: It is hospitality and leisure, some bits of the travel industry.

Dame Angela Eagle: The arts.

Torsten Bell: It is some bits of the arts industries, yes. We need to remember that that is not normal.

Dame Angela Eagle: Aviation.

Torsten Bell: Aviation is still cratered. It is very sectorally unequal. The effect of that is to drive an exceptionally uneven impact across the earnings distribution, very heavily focused on low earners. That is true to a degree with furlough, although the gradient across earnings has softened over time for furlough. On job losses during the crisis, it is very steep. The people who lost their jobs were low earners, basically. That is because furlough was done in a way, for understandable reasons, that supported people with more permanent jobs.

If you look at the types of contracts people had prior to the pandemic, it is a huge determinant of how the crisis affected them. That is true in the UK. We have also done surveys in France and Germany. Germany in particular shows you the same thing. In Germany, they have a bottom of the labour market, which they basically segment from the heavily organised rest of it, and those people have been absolutely hammered. There is a version of that here that you see coming through in zero-hours contracts and the like. They are much more likely to lose their jobs.

Q13            Dame Angela Eagle: To some extent the Coronavirus Job Retention Scheme passed all of them by.

Torsten Bell: The nature of the support, particularly being arranged at short notice, when the crisis is underway, is less well suited. It is not like the Government did not try. Despite what some of the stuff said at the beginning of the crisis, zero-hour contract workers could be furloughed. Agency workers could be furloughed if you had an agency that was interested in a long-term relationship with you, which was not the case in lots of cases, unfortunately.

It is not like the Government did not try, but you were much less likely to be supported. If you were a middle-aged worker who had been with the firm for some time, you were very likely to be furloughed if you were in that position. Job losses, unlike furlough, are very focused on the bottom. The question is then how much that relative insecurity lasts. After the financial crisis, we saw that, when firms returned to hiring, they initially tended to hire in ways that reduced their long-term risk, given the uncertainty of demand. That problem will exist. You choose forms of employment where it is easier for you to manage the risk. We should assume that that is significantly likely to happen in the next phase.

The flipside of that is that you probably get a slightly faster recovery in employment in that world than you would if the firm bought all the risk for uncertain demand.

Q14            Dame Angela Eagle: But you have created more precarious workers.

Torsten Bell: After the financial crisis, we saw that it did not. It started to fall again, but it took a long time. You have to get back to full employment. Then the insecurity starts to fall. Ideally then we would have legislated to regulate some of it out of the system. Given that we did not do that, we should expect a rise to come. If we do not do anything, we need to get to a tight labour market and stay there for a considerable period of time to see it come out.

Dwarfing all that, in terms of low-income households, but not low earners, is what is happening to the benefits system, where changes in benefits will hugely outweigh anything that happens on this.

Q15            Dame Angela Eagle: I am going to come on and ask you about the changes, particularly in the health and social care levy, and how they impact. I wanted to ask one other question. Do you think that we ought to be in a situation where, instead of having a furlough scheme or a shorttime working scheme, put together at speed in the middle of a crisis, we ought to have some change to the benefit system that allows that flexibility in a way that everybody understands, so that we can be more resilient against future shocks?

Torsten Bell: I will try to give a quick answer to that. I have lots of views on that question. First, the nature of the macroeconomy now means that fiscal policy needs to do more of the support of the economy in recessions. I think we all agree about that. Hopefully everyone has learned that from the last 10 years. To do that, you need to lay the groundwork for the policies that do that in the good times, if possible. I remember being back in the Treasury before 2010, when the financial crisis was going on. There were basic things that we would have wanted to do to provide fiscal support to the economy that the IT system said no to. Those problems constrained policy a decade on in exactly the same way, and that is not an okay way to learn from previous crises. Yes, we should change the systems.

Specifically on earnings insurance, which is what the furlough scheme provided, it would be desirable for the welfare state to provide more earnings insurance. That would be desirable in general outside of recessions. If you look at the income falls of people suffering labour market shocks in the UK compared to France or Germany, for example, ours are hugely higher. That is not a desirable thing. We do not offer much for the middle, basically, when they suffer bad labour market shocks. That is not a good situation to be in.

Specifically in a crisis, the ideal from my perspective is that you have built a system for normal times that provides that earnings insurance. It does not need to be a furlough scheme. There is a case for not doing it via the firm and doing it via the welfare state separately, because in normal times you do not want firms hoarding labour and labour not being used productively. If you do it via the welfare state, I would then increase the generosity of that system in the crisis, because you then provide more fiscal support.

When you design these schemes, you want a balance. You want to provide income protection to support the household and the macroeconomy, but you do not want to do it to such an extent that you cause severe work incentive problems. Those relative weights change in a crisis. The importance of the support goes up, and you were not worrying about work incentives when the pandemic was going on, so you could then increase the generosity in the crisis. That is what I would do.

Q16            Dame Angela Eagle: Finally, Torsten, how will the new health and social care levy impact on lowpaid workers?

Torsten Bell: The very lowest paid, who earn below the national insurance threshold, will not be contributing, but big picture, stepping back, most people like me, who come at the tax system from either a technocratic or an economics perspective, would like to see a straightforward rise in income tax, capital gains and dividends across the board.

What you have done is to focus a tax rise on earners rather than people who have other sources of income. The exception is that the Treasury moved on dividends, which is very welcome and does mean that directors of companies will at least contribute more fairly towards it. That is one exception. It means that they are bearing more of their share, particularly the younger ones who will pay this levy for their whole working career rather than, say, two years at the end of it. It is not a good way to go about raising funds, no.

Q17            Chair: Can I give Christopher an opportunity to come in on two points that have just been covered by Torsten there? One is the suggestion by Torsten that some kind of income protection going forward, even if it is not in recessionary or difficult times, is desirable, provided that it still creates, as I would see it, a competitive and fluid labour market. What are your thoughts on that? Secondly, what are your thoughts on the question that Angela asked about the levy, its impact on the lower paid, the dividend tax increase, et cetera? How would you view those two points?

Christopher Snowdon: I would totally agree with Torsten on the levy. I have nothing to add to that. It is fairly straightforward, what is going on. It is not a particularly progressive way to raise money.

On the other point, I do not want to misrepresent what Torsten said. I do not like the idea that the Covid pandemic has taught us how to react to any economic shock. Recessions are just part of the business cycle; businesses go out of business and people lose their jobs. It is a shame for them, but that is what has to happen to weed out bad businesses. It is just the way of life. You cannot try to preserve everyones jobs. I do not think that is what he was saying, though.

Torsten Bell: We are definitely not going to preserve everyones jobs. We did not manage that this time, as the millions of people who have lost their jobs will show. But, no, that is not what you would achieve. No country in the world, even with the most aggressive parttime working schemes, has got close to that objective. You do not want to freeze everyones jobs.

Q18            Chair: It is quite an interesting point. Torsten, if we look at what you are proposing, down to the level of an individual business, what happens? A company comes under some kind of financial stress, and it might have to lay somebody off, reduce their wages or whatever. What does the state do exactly in those circumstances?

Torsten Bell: There are lots of different ways to do this. Your objectives are higher income insurance, particularly for people who are on slightly higher earnings. If you are on a really low income before you lose your job, your level of replacement benefits is still low, but it is higher than for the person who is on a higher income and falls to a floor, because we have chosen as a society to set our income replacement as a cash value rather than a percentage of your previous income, which historically we have done a bit less of, but we have residualised into that form of welfare state.

What do you see happening in countries that have big, institutionalised and sectoral systems, such as Germany, which is what people largely mean when they talk about parttime working schemes? There, where a company is having a temporary tough time, the sector is able to take a judgment about whether that is true or not. That is what then triggers people being able to use these parttime work schemes in normal times: that it is temporary. They do not let you do it forever.

That is harder for us to do, given our complete lack of sectoral infrastructure. We are doing work on this, so I do not want to be categoric about our view. Our prejudice is that the British system is better set up in the short term to operationalise something that is not like a permanent furlough scheme. The furlough scheme operates with basically no infrastructure. It was the right thing to do. We proposed it and we supported it. It has made a huge difference, but you are going to have lots of fraud in the system. That is totally inevitable. We do not have the systems of checks and balance that would mean that I would favour a permanent furlough scheme in normal times.

I would do it via the welfare state with an incomedependent replacement rate for people when they lose their jobs. When the company shrinks in the way you are saying, Chair, people do lose their jobs, but when they go to the jobcentre they are not told, “You are getting £75 a week”. They are instead told, for a certain period of time—it could be three months or six months—“You will get 60% or 50% of your previous earnings while you get yourself back on your feet. It actually does not cost you very much, because Britain has very low and reasonably temporary unemployment rates.

Q19            Chair: It is more about the benefits that would be available when you have lost your job, rather than some kind of furlough mark 2.

Torsten Bell: That would be my starting point, given the nature of our labour market. Other countries would make different choices. If we did have a big sectoral infrastructure, I might favour doing it through firms, but we do not.

Q20            Alison Thewliss: I have some questions about wealth inequality. Starting with Liz, could you describe to us a wee bit how wealth inequality in the UK compares to other countries? What do you feel are the recent trends in wealth inequality? You mentioned some of those in your opening statement.

Liz McKeown: The first thing to say about wealth inequality is that wealth is more unequally distributed than income. If you are looking at income inequalitywe tend to use a measure called the Gini coefficient to do so, because it allows us to look over time and across countries in quite a good way—the Gini coefficient varies from 0%, where everything is equal, to 100%, where it is as unequal as it can be. For wealth, it is 63%; for income, it is 36.3%. You can see that difference. Wealth is just much more unequally distributed.

In terms of what has happened over time, if you are looking at that Gini coefficient measure on income, it has been broadly flat over the last 10 years or so. If you are looking at it for wealth, it has increased slightly in the latest data. You can look at different components of wealth inequality in there. The only component of wealth to have seen a significant reduction in inequality is pension wealth, and that is most likely due to auto-enrolment. Those are the broad trends over time.

In terms of thinking about how we compare internationally on wealth, this is quite difficult at the moment because different countries have been reporting the numbers on different timeframes, which makes those international comparisons a bit harder than otherwise would be the case. The international comparable metric that is generally used in the wealth case is the ratio of mean to median household wealth. The UK figure here is 2.0. It compares significantly lower than the US, which is 7, and it is in the same range as different European countries. Ireland is also 2; Germany is 3.3; Spain, for example, is 2.2. There is some caution in those comparisons at the moment, because of the different times being used.

That is an overview of what has happened to wealth and how it compares to other countries.

Q21            Alison Thewliss: Do you get enough data these days? There is talk of the superrich hiding their wealth and all that kind of stuff. Do you get enough information in order to produce statistical analysis on that?

Liz McKeown: There are always challenges for both income and wealth in the extremes of the distribution and making sure that they are properly captured. The challenges are different at the different ends.

At the bottom end of the distribution, the ability to capture nonhousehold populations is a challenge. There are then challenges at the top end. That is why we look to supplement survey data with administrative data: to give us as good a picture as possible. There continues to be room to improve those measures and to make them more timely as well. Those are big drivers for improvement for us, but they do give a good picture at the moment.

Q22            Alison Thewliss: How much do variations in home ownership contribute to wealth inequalities?

Liz McKeown: Home ownership is an important part. I do not have the exact figures to hand—I would be happy to provide themto give you a quantitative impression of how that compares. Clearly, there are a number of different factors that are determinants of wealth. Age is a very important one, as you can imagine. You see average household wealth increase as you get towards retirement and then drop down again. That is just a profile of age.

Q23            Alison Thewliss: I can see Torsten nodding.

Torsten Bell: There is something that the ONS cannot say, so I can say it. We are very reliant on the Wealth and Asset Survey in the UK to have a clue of what is going on in the wealth distribution. Before that we had all kinds of really awful ways of measuring wealth based on estates data and other things.

That data is imperfect for the reasons that Liz was saying, particularly for the top. We published a paper back at Christmas last year trying to update the Wealth and Asset Survey to do a better job of including the top, which means that wealth inequality would be slightly higher.

The bigger danger is that we are going to go backwards on wealth data. We are relying on Government Departments individually funding the ONS to keep publishing this survey. It looks like we are not going to see the level of support that we need for the survey to continue in as comprehensive a form, and that is a really bad situation. We should be going the other way. We need HMRC to be much more seriously engaging with the ONS to make sure our Wealth and Asset Survey is a good representation, matching in tax data wherever possible. I am definitely worried that we are going the other way. Any pressure that the Committee can bring to bear on the Government decisions taken on the funding of the survey would make a huge difference.

Researchers have spent years saying, “Wealth is more important than it used to be. We need to understand it. It is mad that we are not going to be

Q24            Alison Thewliss: It is difficult to get that detail, and therefore it is difficult to understand.

Torsten Bell: Why would we make the situation even worse?

Q25            Alison Thewliss: Perhaps we underestimate what wealth inequality is if we do not get those figures.

Torsten Bell: The official measures of wealth inequality in the Wealth and Asset Survey almost certainly underestimate the wealth at the top. Yes, that will mean that the wealth inequality measures are lower than they probably would be if you were able to get better data. That is very uncertain, but that is my judgment, yes.

Although wealth inequality is twice as high as income inequality, and Britain stands out for having really high income inequalityin Europe only Bulgaria has a higher level of income inequality—I would go as far as saying that wealth inequality in Britain is low compared to its close comparator countries, insofar as there are good comparator measures.

People think that Sweden is the fairest country in the world; I say this as a pretend Swede. Wealth inequality in Sweden is definitely higher than here. In terms of the industrial mix, everyone wants to say that it is always really great to have loads of familyowned firms. There is a reason why wealth inequality is higher in Germany than it is in the UK, and that is because of familyowned firms. It is not a more equal society. In Sweden it is because they basically protected their nobility after they took all their power off them, and they have not had an inheritance tax in Sweden. That is the deal, but the flipside is higher wealth inequality.

In Britain, if you go through the 20th century, wealth inequality basically falls all the way through the 20th century until the 1980s; then it is flat and then it slightly rises, as Liz says, in recent years. The first half of that is to do with the war killing rich people and estate taxes, duties and income taxes going up at the top. That is what does for the rich in the first half of the 20th century. The second half is about home ownership being spread. Home ownership is much better spread here than in France and Germany, and that is one of the reasons why we have lower wealth inequality.

It is also why, overall, household wealth in Britain is much higher than in those countries. Although those countries have done better than we have on lots of economic metrics recently and they save more than we do, once you start looking at wealth in the round and you see how much wealth British households have in housing, that is the difference. We have all our wealth in housing, and they do not. They spread it. They hold it in cash and in other forms.

The history on wealth inequality and the international comparisons are good news generally. It is still too high, but it is good news. I would not focus on wealth inequality as the right measure to worry about, because the really big deal on wealth is not the inequality changing; it is that the size has got bigger. Wealth used to be three times GDP in the 1980s and now it is seven times GDP. It is growing twice as quickly. The effect of that is not that wealth inequality has gone up; it is that the wealth gaps have risen.

If you are a meritocrat, you should be worried about that, because you should care about the fact that somebodys income cannot see them move from being middle to well off. You basically cannot earn your way to be really wealthy in Britain, if you have these large wealthtoincome ratios. If you are not a meritocrat and you are on the equality side of the debate, you should be worried, because the gaps are now huge. Those gaps have been growing since the 1980s. They have grown again during the crisis. We did some modelling in July showing a further increase of £40,000 in the gap between the average wealth of the household in the top decile by wealth and the middle.

You are not going to save your way to make up for that. This is all to do with asset price changes in existing assets. If you go back 15 years, the economics world was talking about wealth as if it was the accumulation of your income. That is not how the world works.

Q26            Alison Thewliss: What do you do about that, then?

Torsten Bell: That is a huge question.

Q27            Alison Thewliss: Can we have a short answer?

Torsten Bell: Obviously, your tax system should reflect that fact. That is blindingly obvious. What is intended to happen is the opposite, which is that as soon as that increase in wealth starts to hit, in terms of raising some actual revenue, politicians tend to cut the tax. That is the opposite of what we should be doing. The ratio of household wealth to income has gone from three times to seven times since the 1980s, and the amount we raise in wealthrelated taxes is completely flat across that whole period, and that is madness. We need to be dealing with that system.

Secondly, it is not just about the scale of them; they need to be suitable for that. The capital gains tax system actually has to work to do that. At the moment, for example, capital gains is completely ignored once someone passes away, even if they have had huge capital gains on a thing. That makes no sense at all.

As a separate broader policy thing, you need to protect as many areas of life from the effect of that wealth dominating in those spheres, in terms of the education that people can get or in terms of politics. Why is high wealth in the US a problem? It is because they do not protect their politics from the existence of that wealth. If you let that happen, you will quickly find yourself in really big trouble. The same applies to housing and where people live. You need to protect other spheres of life from what you cannot deal with, which is these huge stretches in wealth gaps.

Alison Thewliss: Christopher, what is your take on wealth taxes and wealth inequality?

Christopher Snowdon: I actually have some figures. I dug these out about five years ago, and I cannot remember the source. If you want me to let you know, I can do. This does give the Gini coefficient for wealth inequality, and it really is just much bigger. The Gini in Britain for the last 20 years has been 34%, 35% and 36%. It basically fluctuates around that. Wealth inequality in the UK is nearly 70%. As Torsten says, that is quite low. Denmarks is nearly 90%, slightly above the United States. All the Scandinavian countries, which we associate with low inequality, are certainly higher than the UK.

I would not say that income inequality is especially high in the UK, certainly not internationally. We are not even in the top 100 internationally. It is above average by European standards, but there is definitely more than one country that has higher inequality, although there are different sources. A lot of people use some outofdate stuff from the OECD.

Anyway, be that as it may, if you can manage to ignore the superrich for a moment, partly because there is not a lot you can do about them other than encourage them to leave the country, wealth inequality is really about assets, land and houses. People who own land might be wealthy on paper, but they are not necessarily rich farmers. To take an obvious example, if you are a small farmer you might have £1 million or £2 million-worth of land, but you are not actually rich. You could sell it and go on a cruise, I suppose, but it is your livelihood and your familys livelihood.

It is really all about houses. Wealth inequality is really between the people who have houses and the people who do not have houses. Government policy for many years now has led to horrendous house price inflation, which is just making the situation worse.

I have never really seen a very practical means of bringing wealth taxes in. Land tax is an idea that I find quite attractive. It would be very controversial, but so is taxing peoples houses, as we saw last week. If we are not going to tax blatantly unearned wealth in terms of inflated house prices, which is only going to be handed down to somebody anyway, usually at an age at which they do not need very much money, what are we going to tax? The Government will never do it because of the demographics of the voting system, basically.

Q28            Alison Thewliss: I am conscious of time and I want to move on to let another colleague in, but I just wanted to ask for your thoughts on whether the Governments proposals on social care last week would have any impact on wealth inequality.

Christopher Snowdon: No. It will possibly have some fairly marginal effects on income inequality, but, no, it will have no effect other than making it worse. I am sorry; I was thinking about whether it could improve it. Yes, in itself it will not make it worse, but it will not address the issue at all whereas the incredibly unpopular policy that did for Theresa May in 2017 actually would have done. I thought it was a very good policy. It is the kind of thing that a think tank would come up with, and I suppose that is why people who work for think tanks are not politicians.

Alison Thewliss: I will leave it at that.

Chair: Politics is the art of the possible.

Q29            Rushanara Ali: I have a couple of questions on social mobility and income inequality. Could you take me through whether there is any evidence to suggest that social mobility has stalled in recent years in the UK? Perhaps, Christopher, you can start.

Christopher Snowdon: Yes, this is a very widely misunderstood area. I take my cue from John Goldthorpe at Oxford University, who is really the expert on this. He has become increasingly exasperated as the years have gone on and as he does more and more research. People say that social mobility is going backwards. He says that it has not, but people do not really understand what social mobility is.

Firstly, it should be explained that anything people say about social mobility is at least 25 years old. The nature of the research means that you are always dealing with people who were born quite a long time ago. For a long time, we were looking at comparisons between people born in 1958 and people born in 1970, so people who are now in their 50s. The most recent dataset looks at people who were born in the early 1980s, the youngest of whom was born in 1984.

We have to wait until people are at least into their 30s until we can get any idea of whether they are moving upwards or downwards. We have no idea whatsoever about the social mobility of people who are 18 or in their 20s.

Q30            Rushanara Ali: What is the answer? Has it not stalled or has it stalled, or do we have no idea?

Christopher Snowdon: Based on the youngest people we have looked at, who are in their late 30s, there is still the same amount of fluidity as there has ever been, which is to say that about 75% of people will move up or down by the time they are about 35.

Q31            Rushanara Ali: So 75% will move up the social ladder.

Christopher Snowdon: It is up or down. This is fluidity. Politicians always want to hear about people going up, but, when you have what is often a zerosum game, for somebody to go up somebody else has to go down. For a large part of the 20th century, it stopped being a zerosum game because the working class was shrinking and the middle class was increasing, so there was more room at the top, and therefore more people were upwardly mobile. But there are, tragically, only so many lawyers that a society can need, and so eventually that process has to come to a halt. It leaves more room for people to go down.

Q32            Rushanara Ali: Do you see it becoming more of a zerosum game now rather than expanding?

Christopher Snowdon: Yes, because that can only happen once. The expansion of the middle class can only happen once.

Q33            Rushanara Ali: Unless of course you get economic growth in other spheres, perhaps. It is interesting. Torsten, what do you have to say?

Torsten Bell: I am mainly going to hope it does not only happen once, because it is going to be a pretty grim century if we do not see any economic growth returning at some point, even on that particular one.

Christopher is right. There is a disagreement in the literature. People who approach this from the economic side use the cohort studies and conclude that generally we have seen a slowdown. I should distinguish between different kinds of social mobility here. One is intergenerational, so is your relative position is higher, or lower, than your father, generally? That is what the data allows us to look at, which may not be the perfect measure, but that is what it is in practice. That shows a slight slowdown. As Chris says, the people who tend to focus on more classbased measures, including Johns work, show less change over time. That is one way of approaching it.

Within that space, there are other ways of coming at the same question. For example, it is unambiguously truefrom some very recent data, work we have done, and work LSE has done that shows the same thingthat your parental wealth has become more important in determining your life outcomes, for example your chance of home ownership over time. That is what I meant when I was talking earlier about wealth becoming more important to how your society feels. That is one example. Your parents determine where you live and the kind of house you live in more than your income does.

Q34            Rushanara Ali: Given what you were talking about earlier in relation to savings during Covid, you are seeing an exacerbation of inequality at least between the top and bottom ends and the other groups. Is that going to feed through further or is that not a significant factor?

Torsten Bell: We have been doing a lot of work on that. You see two things going on. There are two changes to household wealth happening from the pandemic, which combine to create the really unusual situation we are in where wealth has gone up despite a huge recession. In most recessions wealth falls because assets fall.

Two things have happened. One is the thing we discussed earlier, which is large increases in savings that are distributionally spread towards higherincome households. Much bigger than that effect is the effect on household wealth of lower interest rates and higher asset prices that follow. The obvious one of those in the UK is housing. In other countries, such as the US, it is also the stock market. For example, our stock market was taking a pounding for other reasons. The UK stock market has not seen the same rise as some other countries. That effect is much larger and is also spread with highwealth households seeing much higher increases in wealth during the pandemic, in really large amounts.

Those gaps are not just about the bottom, by the way. It is the top and the middle. The bottom basically has zero. Adding a lot of per cent to zero makes not a lot of difference, whereas from the middle to the top of these wealth gaps are stretching.

Q35            Rushanara Ali: Liz, unless you have anything further to add on social mobility, I wanted to come on to the next question.

Liz McKeown: I would perhaps just point to two studies that may help address this question in the future. One is a piece of work where we are beginning, in ONS, to look at socioeconomic mobility by different characteristics, including sex, ethnicity, country of birth, religion and limiting health conditions among others. We are going to use our longitudinal study to do that. That is a study of 1% of the census records, so you can imagine that is a very rich data source over time. We are going to see how that plays through across the decades. We are hoping that the results will be available from that next year. That will help really add richness to this.

The other study that is probably worth me flagging is the COSMO study, which the UKRI has funded. It is examining the short, medium and longterm impacts of the pandemic on social mobility and on educational inequality, which is probably an important aspect of this to be considering. That is following the lives of 12,000 school pupils currently in year 11. Although they are not giving answers immediately for the Committee, those longitudinal studies are really going to unpick this in a meaningful way.

Q36            Rushanara Ali: Can each of you talk me through what the drivers of income inequality were prepandemic and where—you have already touched on some of thisyou see the exacerbation of income inequality coming due to pandemicrelated issues? Do not feel that you need to repeat stuff that you have already said, please, because I have a short while to go.

Torsten Bell: The big picture of inequality is a huge rise in the 1980s and into the early 1990s. Big picture, it has been a flat picture since, with a slight rise before the financial crisis, and then a fall largely driven by the labour market hits during the financial crisis and benefit income being protected during the financial crisis and so less affected by that big inflation surge, as I mentioned earlier.

Since then, you see a flat/slow rise during the 2010s. Within that, there are a number of phases. You have bad income growth for everybody in the postfinancial crisis phase. You then get this mini boom either side of the 2015 election where you have decent income growth.

Since 2016, two things have combined to shape the income distribution changes. One is the return of higher inflation and low nominal wage growth because of the postreferendum shock. There is basically no income growth for the two years after that. The second thing impacting that, which gives you the shape of the distribution, is benefit cuts kicking in from around 2016. Those have very materially hit the incomes of the bottom of the income distribution. That is why you are seeing child poverty rise. That is why you see child poverty particularly for large families, with three or more kids, significantly increase. It is why you saw, not in the exact prepandemic year but the two years before that, incomes for the poorest families fall in absolute terms, which is very unusual.

Q37            Rushanara Ali: We are about to see the £20 a week payment being removed. The Resolution Foundations work suggests that relative poverty would increase from 21% to 23% by 202425 and a further 730,000 children would fall into poverty. What is your take on that? If the Government were—we live in hopeever interested in reducing income inequality and reducing the impact of that, what should they be doing?

Torsten Bell: “Not that is the blindingly obvious answer. It is a huge cut to the incomes of the poorest families in the country. It is a 5% cut on average, but a million households are losing 10% of their income overnight in October. That is not what you should be doing. It is not good for the macroeconomics, for the reasons we have been discussing earlier, in a highinflation phase with uncertainty about the trajectory for the recovery, even though there has been good news in general over the last six months. It does not make sense on the macroeconomics.

The introduction of this extra benefit spending during the pandemic by the Government was very welcome. Poverty almost certainly fell in the crisis. Both absolute poverty and relative poverty almost certainly fell. We will have to wait for some data from DWP and ONS to confirm that, but it is basically almost arithmetically impossible for that not to be true. That was very welcome indeed, and it helped a lot of people through some really tough times. We are about to reverse a central plank of that at a time when inflation is at a high level, and that is not what we should be doing.

For the bottom of the income distribution, there is probably nothing else that will happen this Parliament that can have as big an effect on the income of the bottom as this change. We have done some nowcasting and forecasting for income growth using the OBRs projections across this Parliament. This change probably halves the amount of income growth that the bottom will see over the course of those years. It moves that from being broadly flat across the income distribution to being a regressive inequalityrising Parliament, with child poverty in particular rising.

Q38            Rushanara Ali: It is not really levelling up, then, is it?

Torsten Bell: On the income distribution, it is definitely levelling down, yes.

Liz McKeown: I will perhaps pick up on how we look at the effects of tax and benefits on income inequality. Each year we publish a publication on the effects of tax and benefits, which takes us through what effects they have on income inequality.

If we take the last year for which we have data to illustrate this, 201920, we start at the point of original income. This is all sources of income from employment, private pensions, investments and other nonGovernment sources before any tax and benefits. The Gini coefficient that we have been talking about at that point is 50.5%. Once cash benefits are considered, that reduces to 40.8%, and it reduces further to the 36.3% that we were talking about earlier once taxes and benefits in kind such as public education and healthcare are factored in.

We can look over time at the gap between the original and the gross income Gini coefficients. That gap has reduced over time, which is likely to highlight the diminishing effectiveness of cash benefits at reducing income inequality over time. It reflects the moderation in value of cash benefits received relative to households original incomes. That publication next year will highlight how the tax and benefits changes that we are seeing at the moment play through.

Christopher Snowdon: The story for all these measures, really, whether it is child poverty, relative poverty, Gini inequality or any other measure of inequality, for the last 20 years has been really quite dull and nothing much has happened. They have moved within a very narrow band.

Q39            Rushanara Ali: Why is it dull? Is it because you do not

Christopher Snowdon: Because it is a straight line virtually and nothing ever happens, despite the fact that we are, particularly over the last 10 years

Q40            Rushanara Ali: What does that mean? What do you mean by “nothing ever happens”?

Christopher Snowdon: I mean statistically. That is all I mean.

Q41            Rushanara Ali: How do you definenothing ever happens when we are talking about quite dramatic shifts in terms of peoples income due to certain policy changes? I do not really follow. I do not really understand what you are trying to say.

Christopher Snowdon: Take relative poverty, for example. That is 22% currently; 18 years ago it was 22%; it dropped briefly to 21%. It has never changed from between 21% and 22%. This is at a time where for at least—

Q42            Rushanara Ali: What do you make of this 730,000 figure? I am sorry; I am pitting you both against each other now. What would you say to the Resolution Foundations figure of 730,000 children falling into poverty because of just one policy change? Is that nothing ever happens”?

Christopher Snowdon: It is the boy who cried wolf.

Rushanara Ali: Do you disagree with it?

Christopher Snowdon: It is unlikely, because I am so used to think tanks claiming that we are going to see spiralling inequality, spiralling child poverty or X number of people falling into poverty on the basis of usually quite minor changes to the benefit system, austerity or whatever it is. We have had these constant predictions of doom for 10 years.

The Equality and Human Rights Commission predicted a child poverty rate of 43% by next year a few years ago. I have a bet with Jonathan Portes that this will not happen. He wrote the report. It clearly is not going to happen. What is it at the moment?

Q43            Rushanara Ali: I represent a constituency with very high levels of poverty in reality. Are you telling me that I am imagining it?

Christopher Snowdon: No. What I am saying is

Q44            Rushanara Ali: Are you saying that, because you disagree with it, it does not exist?

Christopher Snowdon: I am saying that the proportion has not changed over time despite the fact that people are constantly saying, “Well, next time it is going to shift.

Q45            Rushanara Ali: But it has, actually.

Christopher Snowdon: Relative child poverty after housing cost has been around about 29% for about 20 years. Last year, 201920, it did tick up from 29% to 31%, which is about the same as it was in 2007. It has not come anywhere close to 43%, which would be breaking all records by a considerable margin, which is what people have claimed. It has not risen significantly in any other year—

Q46            Rushanara Ali: It is 60% in my constituency, and it was not 60% a few years ago. Torsten, you were trying to come in.

Torsten Bell: Income inequality has not moved up and down hugely. Poverty on a societywide level peaks in the 1990s and does then fall. Remember that smallish percentage movements here are lots of human beings. Let us be a bit careful. Poverty is between 24% and 25% in the 1990s, and it falls down to about 22% or 21% in the 2000s. Within that, though, real things are going on. Pension poverty is falling; child poverty is not.

If you look at child poverty, I definitely do not agree that big changes have not happened. That is, by the way, despite the fact that our official data hides how big those changes are for reasons that we probably do not want to go into. Hopefully the ONS and DWP are going to fix that in the coming years. Child poverty falls really significantly in the early 2000s. It then rises just before the financial crisis. It does fall during the financial crisis, again because of the benefit protection that I mentioned earlier. It then has risen by about 700,000 children since the bottom of the shortterm child poverty trough in 2011.

As I say, another 700,000 could go into child poverty, which would take us back to those 1990s levels. That is a long way short of the 43-odd per cent that Christopher was talking about, and I have not seen the paper he is referring to. Our anxiety is that big increases have happened, and they have happened in the recent past, the last five years. Could they happen again? If we go ahead with the £20 cut, compared to what poverty levels were during the pandemic, when, as I say, poverty fell because of what the Government did, it is almost impossible for poverty not to rise in the coming years.

Rushanara Ali: You said “1990s level”.

Torsten Bell: Yes, the 1990s is the peak for child poverty.

Rushanara Ali: That is staggering, really. It is setting back progress achieved by successive Governments in different ways.

Q47            Chair: Can we zero in on that for a second just so I am absolutely clear? Clearly, if you have a benefit arrangement where 6 million people are receiving about £1,000 pounds a year et cetera and you take that away, that is certainly going to have an impact—it is hard to imagine that is noton various measures of poverty. Are you saying that the impact, depending on which measure you look at, would take poverty levels back to or worse than they were before that UC increase was brought in?

Torsten Bell: Introducing the £20 and taking it away again clearly will net out in terms of its effect in the future. What is going on is that another load of benefit cuts are happening over time underneath that. Remember that what is unusual about the benefit cut happening in two weeks time is that it is, first, huge and, secondly, happening overnight. That is really unusual for a large benefit cut.

Historically, benefit cuts are done either to new claimants or new children in a family—that is what George Osborne did with the family element and the twochild limit—or they are freezes on uprating, so you cannot see them happening and they take years to come into effect. The things that are working their way through the system at the moment are the freeze that was in place prepandemic over four years, the twochild limit and the family element cut. Those will still be being implemented in 2030, by the way. This is a long implementation period, because children take decades to grow up.

This is going to be feeding through into poorer families household incomes, and it will be pushing down income. Last year, the £20 pushed against that. When you take it away, you remove the protection; you remove the effect that had on softening what was happening to poverty rates, if that makes sense.

Q48            Chair: Does that take poverty rates back to being worse than they were prior to the UC uplift coming in?

Torsten Bell: Modelling does not allow us to have that kind of precision. To take us above the 1990s would be quite hard, but taking us back towards the 1990s levels is eminently possible, yes.

Q49            Anthony Browne: My questions are going to be about regional inequality. First of all, to understand better the causes of it, there are various reports that say we have some of the widest regional inequality of any G7 country. Why is that? Then I will be asking about the impact of the pandemic.

I do not have any particular preference in terms of order. Torsten, you seem to speak first.

Torsten Bell: Why doesn’t someone else do that, then?

Q50            Anthony Browne: Christopher, do you have any views?

Christopher Snowdon: I do not. I have not done any research on regional inequality, so I will not get involved.

Liz McKeown: It is right to say that the UKs regional or spatial disparities in productivity are large in an absolute sense and in comparison with other developed countries. A lot of that is driven by the fact that London has some of the highest productivity of almost any region, when you are looking across Europe. Paris is another example there.

When you are looking at these differences, it is important to look within regions as well as between them. You see some really quite marked variations at that lower geographical level as well, which is important to consider.

What might be driving some of those? In principle, there are a number of factors that could be playing a part here. You have to look at the workforce. In human capital terms, what are their skill levels and their health levels? Those are going to be important. You need to look at what productive assets and infrastructure there is within the region. You have to look at it from an institutional point of view at a local level. It is also important to look at the composition of economic activity in the region.

In practice, high-productivity regions outperform low-productivity regions across all those dimensions. You could take any one of them and you would see those differences occur. That is probably a broad answer to your question.

Q51            Anthony Browne: I am particularly interested in whether you have an answer to the question about why it is greater in the UK than in other countries like Germany, for example.

Torsten Bell: There are different ways of thinking about these regional gaps, and they give you very different answers. It is really important that we distinguish between them. The productivity gaps in Britain are large. There is a large row about how large they are, depending on which regional size you look at, but a consensus from all of us is that they are large.

They particularly rose in the 1990s. Although we read Victorian literature, and we talk about the northsouth divides and things, it is not true that our regional gaps are ever growing. The regional gaps fall in postwar Britain, and they only start rising again when we get into the late 1980s and 1990s. You then do get a rise in the 2000s, which is particularly about the top moving away. We have not actually seen productivity gaps rise in the last decade, but they have not come down either. They have stayed stubbornly high. That is on productivity.

On household incomeand, again, people do not say this enoughgaps have been falling between places pretty consistently now. We are probably now back down to the lowest variation in income across Britain that we have seen since the 1970s. It is moving the opposite direction.

Q52            Anthony Browne: What is causing that?

Torsten Bell: What is causing that is us eventually closing the employment gaps that the 1980s gave us. The effects of the 1980s recession, because all recessions are different

Q53            Anthony Browne: This is after deindustrialisation.

Torsten Bell: Correct, there were hugely unequal unemployment effects that stayed with us for a generation. That has happened now either for good reasons, with areas catching up, or for bad reasons, because some people have now passed away or exited the labour market. Employment gaps have shrunk hugely. That was happening before the financial crisis, but it happened a lot in the last 10 years. Under this Government, you have seen big falls in employment variation.

There are two other things going on. You are also seeing a fall in the earnings gaps between places, which is partly the minimum wage, because it is set nationally, pushing poorer and lowerearning places up towards the rest. More generally, we do seem to be seeing a closing of earnings gaps between areas. Those are the two bits of good news.

The bad news that is closing income gaps is that housing costs have been becoming stretched. The way to think about this is that housing costs have risen most, and they were already higher, in higherincome areas and higherproductivity areas.

Q54            Anthony Browne: This is particularly London and the south-east.

Torsten Bell: London is the classic example, but it is true if you look at both ends of the A64 in Yorkshire. It used to make a lot of financial sense to move from Scarborough to Leeds or York. It does not make much financial sense now, because, by making that move, your rent would go up by more than your earnings, even though the productivity gap has not closed. These are really important things for understanding why, for example, young people these days are less likely to move from a lowproductivity area to a high-productivity area.

If you were a 1980s Thatcherite and you wanted economic dynamism to come from people going to the places where there are opportunities, the housing costs differential—it is not the levels; it is the differential—means that it does not make a lot of sense for you to do it. If you look at where people move to these days, they are more likely to move to lowerhousingcost areas than they used to be and less likely to move to higher-housingcost areas.

All I am saying is that income is doing a very different thing to productivity. Decide what you care about, and do not accidentally mess up the income gaps closing in your attempt to close the productivity gaps. Policymakers should care about both. We have the levellingup White Paper coming in a few weeks, hopefully. If you were looking at what I have just said, I would have thought that you would conclude that you want to do something about your housing cost differentials and your productivity gaps.

To answer your question on Germany, what is going on? There is historically more variation in regional strength in the first place. Secondly, there is a huge national policy impulse from reunification, which saw a national mission to close this gap because of the huge eastwest productivity divergence, in which staggering sums of money have been spent. There were also things that we would see as deeply illiberal—well, not deeply illiberal, but I suspect politicians here would not dowhich were basically to encourage some areas to shrink and other areas to grow. I suspect we are not happy to do that. We have an advantage; we are a smaller country.

Q55            Anthony Browne: We want to level up rather than level down.

Torsten Bell: I am not even sure what it is in levelling terms. It is definitely shrinking some poorer places, yes. Some people may have thought that was fashionable in the past, but that is not what politics in Britain is about these days. It is less important in Britain as a way forward.

In the US, if you are a small town in the middle of the Midwest, you can understand that population change probably plays a bigger part because you cannot connect to any other economic areas if your area is failing. England in particular is a very, very small country. Even the economic determinists, who think you cannot turn around these longterm productivity gaps in general, should be more optimistic about Britain. There are not that many places that cannot be connected to the 21st century economy. It is more difficult in some very large countries.

Q56            Anthony Browne: I want to ask now about the impact of the pandemic. I have two questions, really. First, is there any measurable impact on regional disparities as a result of the pandemic? London is big on hospitality, and that was one of the sectors that got hit while manufacturing was less hit generally. Also, we have all learned how to work from home and work on Zoom. There might be a shortterm impact from that, but what might the longterm impact be from that? Would that help remove regional inequalities?

Liz McKeown: The labour market is a good place to start looking at this question. At the regional level, over the course of the pandemic we saw a decline in the number of payrolled employees across every region, but the initial fall was more substantial in London than in any other region.

Q57            Anthony Browne: That is because of hospitality, presumably.

Liz McKeown: Exactly, yes, and the effects of tourism being substantial in London. You drew attention to working from home. We saw that London and the surrounding areas to London having the highest rates of working from home in 2020. Around 43% worked from home at some point in the last year in London, which is really quite significant. That again will have had an impact on both commuting patterns and displacement of activity.

What we are seeing in London, though, is a bigger growth in employment recently, although it is still, alongside Scotland and the south-east, one of the three areas where the number of payrolled employees is below preCovid levels. London is furthest behind, but it had the largest increase in the last month.

Like some other areas, the regional picture will take some time to evolve now. We need to keep a careful eye on the data as it comes through, because we did see this initial pattern. It was really quite different in London, but we are seeing the recovery and how that now plays through in a different way as well. We are not in a settled position yet in terms of peoples future patterns of where they are working from and mobility. That is one to keep an eye on.

Q58            Anthony Browne: I will ask Torsten the same questions. What has been the impact of the pandemic on the regional variations? Is working from home going to have a longterm impact on regional disparities? People can work for companies in London even though they are based in the north of England.

Torsten Bell: I broadly agree with what Liz has just set out. The only caveat I would add is that the nature of how the pandemic had geographical effects was partly connected to the point you were making now about the sectoral mix. All areas have the sectors we are talking about. That is why it was a more broadbased recession than most. This is not a 1980sstyle recession for that reason.

In terms of geography, I would focus slightly more on thinking about the areas where people had to travel to spend, where the demand was not homegrown. Those were the areas that were hardest hit. Two obvious ones stand out: city centres and tourist hotspots. In the first peak of the hit, those were the areas that were hit hardest. You can see that in the furlough data; you can see it in the employment data; you can see it in a lot of ad hoc survey data on spending.

The recovery pattern, though, for those different groups is different. Areas where the lack of domestic tourism was the problem bounced back very fast. That is like South Lakeland or Devon. It appears that everyone apart from me was in Cornwall this summer.

Anthony Browne: I was not.

Torsten Bell: That is lucky. Your chance of having Covid is materially lower as a result of not being in Cornwall over the summer, but everyone else was.

You see very fast improvements in employee numbers in the touristreliant areas. You see less quick bounceback in cities, in particular in London. Two things mark London out from other cities. It has not actually that it has more hospitality and leisure; it is that that is driven by foreign tourism more than other cities. That has not bounced back, which is why you also see Slough, the places near Heathrow and Crawley, for Sussex and Gatwick, hugely hard hit and still very hard hit.

It is the patterns of the recovery that are driving the geography. Eight out of the 10 highest furlough rates are still in London. London is the basketcase in the recovery from this crisis. We are levelling down very substantially in one way at the moment. It is really significant.

How does it last? How does working from home play in? First, I should say I do not know. There are lots of surveys being done now about what peoples expectations are. They all tell you what you would have heard in the pub. Again, be careful here: we mean professionals. Where are professionals going to do work from home? Half the population is not going to do any working from home, because they go and actually do a proper job. But those of us who can work from home will probably be doing three days a week or four days a week. That is what the surveys show. If it helps, that is what the Resolution Foundation will be doing.

If that happens, thinking about the structural change that that will drive, it is credible that it will have an effect within regions. Some people may choose to commute slightly further for three days a week than they would have done for five days a week. It will change spending patterns, even if nobody moves. There will be less spending in city centres and more spending in local high streets. There are some studies now trying to quantify that.

At one level, the change in working from home is very large, and it will have an effect on employment, but the effect on employment means that professionals will get more flexibility in where they work. Those who service the demand that professionals drive, instead of commuting to city centres, which usually means coming into the pie, if it is a city, will travel laterally around the city. I live in Haringey. Lots of workers in Haringey are currently—this is why there are high furlough rates—servicing office blocks in the middle of London. They will spend more time going across to Hampstead or to other rich parts of London. That is a big structural change for them. The absolute size of it within the grand scale of the UK labour market is small, but it will be significant for those people.

I have read slightly too many newspaper columns saying that the work from home revolution is going to solve the UKs regional productivity gaps. As you say, someone is going to be able to live in Richmondwell, Richmond actually has quite high income already—or North Yorkshire and have a job in London that they never actually turn up to. Yes, maybe in small numbers, but let us not get carried away yet.

Anthony Browne: You are probably right. This is absolutely fascinating and I would love to carry on asking you lots of questions, but I will get told off by the Chair, so I had better hand over.

Chair: No, never.

Christopher Snowdon: I am sorry. Can I just say something about the regional inequalities?

Chair: Yes, please.

Christopher Snowdon: The work from home thing is a very positive development, but how it affects inequality I do not know. I am afraid I do not see everything through the lens of inequality. Through the lens of efficiency, it is a very good thing. Excessive rents and unnecessary commutes are a simple waste of money, and the less of it we have, the better.

Rents are now falling in London. It is not just that growth is declining; they are actually falling. It is extraordinary.

Anthony Browne: You mean commercial rents.

Christopher Snowdon: No, domestic rents. The figures were just out today from the ONS. Unfortunately house prices are still going up, although not as quickly as they did before. If people can get out of London, live elsewhere, not waste money on commutes and get a cheaper property, that is a very good thing for the economy.

I guess there will be marginal effects. As Torsten says, you will have more wealthy people living elsewhere and spending their money, yes. There will be some kind of knockon effect there, but it is not really going to transform anywhere in and of itself. It is a good thing, and it should absolutely be encouraged.

A few months ago, MPs were complaining about Pret A Manger missing out on sandwich sales in the middle of London. Who cares? Who cares about Pret A Manger? This is creative destruction. It is what I mentioned before. It is very sad for people working in Pret A Manger, but it is very good for people working in the sandwich shop in Brighton or wherever these people have moved to. It more than evens out. It is a good thing.

Q59            Siobhain McDonagh: I just want to look at intergenerational inequality. I apologise if that is not what you are interested in, but unfortunately that is what today is about. My questions will just look at underlying inequalities between generations. Millennials born in the 1980s are the first postwar cohort to have lower incomes during early adulthood than the cohort before them. For my first question, I would like to look at housing.

In 1997, 55% of young people aged 25 to 34 owned their own home. By 2017, it had fallen to just 35%. Meanwhile, one in 10 adults owns a second home while four in 10 do not own their first. When it comes to owning your own home, intergenerational inequality could not be starker. Torsten and Christopher, why do you think such a clear intergenerational divide in home ownership exists?

Christopher Snowdon: Because the price of houses has gone up massively since the boomers bought them. It is as simple as that. We have not been building anywhere near as much. I mentioned before the idea of taking the money out of housing to go into social care, inheritance tax or whatever. There are things you could do, in principle and in practice, in terms of clawing some of this unearned wealth back.

Preferably, we could build a lot more houses. We would only need to build on 2% of the green belt to solve this problem.

Q60            Siobhain McDonagh: I did an event at the IEA on this, around two years ago, about building on the ungreen green belt.

Christopher Snowdon: Yes, it is not green for the most part. There are huge bits of land around tube stations, train stations and so on, which could be built on.

Siobhain McDonagh: There could be a million houses, yes. I even went to the Conservative Party conference and did one of the fringe meetings.

Christopher Snowdon: It needs to be done, but again democracy is getting in the way of it, because the nimbies, the boomers and the people who want to protect their wealth have a disproportionate say in the matter, as it were.

Torsten Bell: It is a little more complicated. I agree on the policy recommendation that it would be desirable to build more houses, particularly in some areas where you can see overcrowding rising. Prepandemic, anyway, London was the standout case study of that, although there are other examples of it.

That would definitely help reduce housing costs. House prices are driven by two things, though. They are driven by housing costs, so rents, and by interest rates. The structural fall in interest rates since the 1980s is the principal driver of the rise in house prices, although the lack of supply is a serious contributor and definitely helps explain why some areas have seen bigger rises and falls than others.

If all you care about is ownership rather than housing costs, we are not going to succeed in building our way out of it, so we need to think more broadly. The reason why interest rates are driving that is that, to state of the obvious, your mortgage payments need to equalise in the end to the cost of renting, because that is the alternative you have. When interest rates fall, house prices rise. This is getting techy, but in your mortgage payments your capital payment increases and your interest payment falls, and they roughly stay the same in aggregate. That is the economics of what is driving up house prices and what is driving up all the other nonsense assets that you hear about on the news

Around the world, why are asset prices going up? Lower interest rates over 30 years is the reason. When that happens, it makes the deposit a bigger barrier to home ownership, because of the relationship between your income, your ability to save, and the deposit you need. The deposit is set as a percentage of the house price; it is not set relative to how much it is going to cost to service that, for good reasons to do with the stability of the financial sector. That is the thing that is driving down young peoples home ownership, and that is why it is hard. If it was just building it would be easier, although, as Christopher says, it is really hard because of the politics.

What is that doing, then, in practice? That is the economics that is giving you boomers who have access to cash becoming second homeowners and young people who do not have access to savings, because of the life cycle, being unable to get on the housing ladder. The figures are stark. I would actually say, by the way, that nobody started talking about falling home ownership until the early 2000s. Youth home ownership starts falling in 1989, and no one notices for well over a decade and a half because aggregate home ownership did not start falling, because some boomers were still getting on the housing ladder in their 50s and 60s.

We did not notice it for 15 years, and I am including myself in this. Policymakers were asleep at the wheel. The problem was that we then did not get to do anything about it. If you want to do something about it, you are going to need to build, yes. You are going to need to take a much stronger stance on whose demand for housing you want to favour.

You are going to have to say in the tax system, “We are going to support the demand for your first home, your primary home, and we are going to penalise other forms of ownership”. Otherwise you are not going to be able to counteract the fact that the buying power, because of this deposit barrier, is very different between these two kinds of groups. You can do that in the tax system. Some countries are looking at doing that via regulatory methods. This rebalancing of demand is the thing that you are going to have to get at, if you want to really make a difference.

Alongside that, you should recognise that this problem is not going away unless rates go away. Among the reasons why you need to build social housing is that you need to ensure people have secure tenures. It is not the only good thing, but one of the best things about home ownership is the security of tenure. You are not susceptible to changes in house prices in that area, meaning that you cannot live in the place you are used to living in.

Other forms of tenure could provide that. They cannot provide the enforced saving, which is also a big benefit of home ownership, but they could provide other ones. You can definitely do that in the social sector, and we have stopped doing that recently. You could do that in the private rented sector, but it is harder.

Q61            Siobhain McDonagh: Liz, are the gaps widening? What information do you have?

Liz McKeown: Looking at intergenerational income inequality is something we did for the first time with our publication on the effects of tax and benefits in 2019, where we looked at how peoples income depended on what generation they were born in and how that appeared at different stages of their lives. In that analysis, we did see that, if you look at people who were born in the 1930s, 1940s and 1950s, they have higher average incomes than the preceding generation at every stage of their lives. Then you see that pattern break down and households with heads born in the 1980s have broadly comparable median incomes to those born in the previous decade, and the results for those in the 1990s were inconclusive. You do see that sort of intergenerational inequality playing through.

There are some challenges to that analysis, and there will be as we look to do more of that going forward. When you think about household income, you now are seeing a pattern of younger people being more likely to be living back at home with their parents. That increases their household income, which, when you are making some of these comparisons, makes it quite difficult to play through.

Thinking more broadly in terms of what impact the pandemic has had here, there have been specific negative impacts for younger people. We have already talked about the fact that the sectors they worked in were more likely to be impacted and therefore there were negative impacts on the labour market side.

We also see that playing through on the householdfinance side. For example, we found a significantly higher proportion of young people reporting that the pandemic was affecting their household finances. I think 30% of young people reported that compared to only 13% of older people. Of the young people who did report that, 84% said that they had had a reduction in income and 38% said they were unable to save. Across the labour market, across household finances and across financial resilience, we were seeing that young people had these negative impacts during the pandemic period.

Q62            Siobhain McDonagh: The Nuffield Foundation announced yesterday that a staggering 36% of children in families with a child under five are living in poverty. They pointed to a number of factors, including changes to the benefit system, the rise in insecure work and the growth of the private rental market. When it comes to private renting, people in their mid30s to mid40s are now three times more likely to privately rent than 20 years ago. Torsten and Christopher, what impact do you think the continuing rise in private renting will have on intergenerational inequality?

Christopher Snowdon: On wealth inequality, it will have a very significant effect, because obviously they will not have a house to pass down to their children. I would say, however, that the real question is housing costs.

I appreciate that most people would probably rather own their own house than rent, but, if you build a lot of houses, even if you get into the situation that Torsten is getting at, where it is swamped by buytolet boomers or large commercial investors, you still reduce the price of rent. You are still flooding the market with supply, so you would at least get rents down, which would help a great deal, even if you do not necessarily get house prices down. As Torsten mentions, there are possibly things that you could do to dissuade the buytolet market, if you wanted to do that. They are not very free market policies, but you could look in that direction.

In direct answer to your question, yes, it is fairly clear that, if you have a society in which there are lots of people renting and not putting anything away for a rainy day, particularly in a market where house prices keep going up and show no sign of ever stopping, you are going to see wealth inequality get worse.

Torsten Bell: I will break that down into a number of bits. There has been a shift in the last 10 years towards the private rented sector. For young families, that is coming from both ownership and less access to social housing. Social housing has become less good at being available to young people. If you go back 30 or 40 years, almost no children were brought up in the private rented sector. We now see quite a significant number of children brought up in that tenure. There have been very large rises over that time period, but it is for both reasons. It is not just the home ownership one.

The effects of that are, first, lower security. We have set up our private rented sector to offer shortterm rents for a short period of your life, not for bringing up your kids, where you need security, because you need to be near the school, for example. That is about the regulation of the sector, but the security is the big problem.

On the housing costs, the rise in child poverty is being driven by the switch in tenure. In general, the housing tenure we have moved into is a higher housing cost tenure. It is also because the change over the last 20 years is to push down the housing costs of richer households, because they come down with interest rates, but people who have to rent do not see their housing costs fall with interest rates. Instead, what they have seen is a slight rise particularly in some areas over time. It is not huge, but it is the compositional shift into the private rented sector that is the main driver of the higher housing costs.

The other contributing factor is that benefits have become less generous specifically for young families over time. We did some analysis published at the end of August, which takes all the benefit changes since 2010 and sees what the impact has been across different age groups over that time period. The biggest falls are among children. Someone born post-2016 will see their household benefit income in real terms come down to £93 a week on average compared to £118 a week for the generation before them, which is Gen Z. I do not think the ones born post-2016 have a name yet, but whatever they are called.

We have chosen to make the benefits system less generous, particularly for those with children, while we have made it more generous for pensioners during that same time period. It is those changes collectively that are giving you higher child poverty rates, particularly for large families.

Q63            Siobhain McDonagh: Liz, how do you anticipate that rises in private rent and the fall in home ownership will impact inheritance in the coming decades?

Liz McKeown: I must say we have not made an analysis of the link between that and inheritance in the future. We could certainly come back with the relevant statistics that we do have in that space.

Q64            Siobhain McDonagh: I highly recommend to everybody here to take a minute to listen to the outstanding speech by my colleague the Right Honourable Member for Barking, Margaret Hodge, in yesterdays debate on the Governments national insurance hike. Her speech highlighted clearly the intergenerational injustice that will be caused. “The young will pay for the old. The struggling tenant will pay for the wealthy landlord. The assetpoor worker will pay for the assetrich retiree. Christopher, is she right?

Christopher Snowdon: Yes. I do not know if I would get quite so excited about that one particular tax. It is one of many taxes that we are probably going to see over the coming years. What is it going to raise£12.5 billion a year? Net zero, according to the Governments figures, is going to be £50 billion a year. That is a Government figure, so you can probably treble it. You are going to need four taxes like that just to pay for that. You have the NHS, which is going to continue demanding more and more money. Social care is not even really going to be funded from this new tax. That will need something else.

We are in a hightax, lowgrowth country. Ideas for what to do about it have been bounced around already today, but they are all politically impossible. We need to accept that sadly. Inequality has not changed for 20 years. I do not see it particularly changing in the near future with or without the effect of the pandemic.

Most people would like to see tax being broadly progressive, but the reality is that there is just a limit to how much tax people are prepared to pay. People have been calling for more funding for the NHS for a very long time, but as soon as they have to pay for just a bit of it they complain. It is the old problem. People want higher public spending and lower taxation, and they cannot get it except through debt, which is why we have £2.2 trillion worth of debt as well.

Siobhain McDonagh: Thank you, Christopher. I am sorry to Torsten and Liz. I have some great questions for you, but my time has run out.

Chair: We are expecting a Division in about 15 minutes. We have two contributors left. I do not want to put pressure on anybody to curtail the time that they have, but just bear that in mind. We might just get through before the Division.

Q65            Julie Marson: Let us canter through this. I would like to look at the impact of the crisis on women. I wonder if you, Liz, could give us a brief picture of what you have found in terms of both absolute employment statistics and some of those broader areas where women have felt the impact of the crisis.

Liz McKeown: It is worth looking at this broadly. If we look at the labour market itself, men were more impacted than women. That could be because women were more likely to be working in the public sector, for instance.

We looked more broadly at the impacts of the pandemic on women and men, and it is worth doing so across a number of dimensions. In health terms, in mortality terms, men were more adversely affected than women by the pandemic. In wellbeing terms—there are a range of measures there, looking at things like loneliness, depression and personal wellbeingwomen had more of a negative hit from the pandemic than men did.

The other interesting thing to look at is other types of work during the pandemic, and in particular unpaid work in the home and with childcare. We saw there that women were spending 44% more time on unpaid household work and 55% more time on unpaid childcare than men in the first part of the pandemic. We are looking at the March and April 2020 period. This was from our Time Use Survey. When we repeated it again in the September and October period, it had increased. At that point, women were spending 64% more time on unpaid household work and 99% more time on unpaid childcare than men.

If you want to compare the experiences of men and women throughout the pandemic, it is important to look across the labour market, the health impacts and what it has meant for their contribution to unpaid work as well.

Julie Marson: Christopher, have you looked at this particular aspect of the impact of the pandemic?

Christopher Snowdon: No, I have only read the ONS stuff, so there is no point in me repeating it.

Julie Marson: Torsten, do you have any comments on this?

Torsten Bell: We have done some papers on this area. The big picture that Liz has set out is basically consistent across all the different research in this area.

To dig into what is going on a bit more, in the labour market the gender split is explained by the publicsector point that Liz made, which is that public sector workers were more protected in terms of losing their jobs. They may have been at a health risk, but they did not lose their jobs. The disproportionate selfemployment of men is also a big contributing factor, particularly the kinds of selfemployment that saw big falls. Even today, 18 months in, you can see that in the male selfemployment figures being down a lot.

From the work that we have done, you can see the parenting gap in the hours worked. You can see it in hours worked in the first lockdown, where basically mothers working hours came down a lot more than fathers. Forget men and women for a second; that is specifically within parents. That gap was less severe in the second lockdown. I do not know whether that is people learning their lesson, people not putting up with it anymore or whatever was going on, but that did close. Even then, the survey evidence showed that mothers were much more likely to say they were changing how they were working than fathers were. We can say that is a consistent factor all the way through.

I completely agree with the point Liz is raising about the mental health gaps being very large. They were probably twice as large for women as they were for men in the crisis, which is big. We see a gap anyway in steadystate data, but that is a very large difference. We will see whether the labour supply shock to men, or the damage to peoples wellbeing and mental health, which has been worse for women, is the lasting effect.

The main thing I would say is that everyone is desperate to say that it is worse for men or women. The big picture that we see from all the data is that it was actually quite different. The main thing is that it was just different. There were different experiences, and that is probably more how we should reflect on this.

Q66            Julie Marson: That is what is interesting. Some of those differences are quite baked in culturally with those responsibilities that women take. What does that tell us about longerterm effects of unemployment even on the gender pay gap, for example?

Torsten Bell: I definitely do not have all the answers to that, because that is very uncertain. I have some thoughts.

First, one of the reasons that we have seen less of a labour market effect for women—this does reflect a longerterm social trend—is that we are seeing some women going into work and looking for work when they might not otherwise have done so. Given our existing society, women are disproportionately likely to be the second earner in a coupleless likely than they used to be, but still more likely. What happens is that, as the household income takes a shock, which is what happens in crises, we see that the margin of adjustment of doing more work falls on the woman.

That is what we saw after the financial crisis: more employment of women in couples, and big rises, particularly older women approaching retirement. You also saw those who were working doing more hours. The women in work during this crisis actually increased their hours, even though the hours overall have come down a lot.

You are seeing a labour supply increase among women that is continuing the pattern of the 2010s, although it has sped up a bit during the crisis. If that lasts, it might counteract some of the labour supply reductions that we might see postcrisis. Migration is one of them, but so might be early retirement and people who have a bad experience. Long Covid may easily be a labour supply shock to us about which we have no idea. The evidence is all over the place on how big that is going to be. We could expect some of that to last unless we see really fast wage gains suddenly coming through the system in the future.

We tend to see this pattern of men doing worse in terms of job losses after recession. We see that in every recession. If anything, it is slightly less marked towards men this time than normal, even though it is still disproportionately towards men. We have, after some recessions, seen that lead to spikes in longerterm male unemployment. It might not be as bad as last time, but it could happen.

Christopher Snowdon: I would just say that, if many more women had died of Covid than men, and if more women had lost their jobs than men, nobody would be claiming for a single second that women had not had it worse than men.

Q67            Julie Marson: Liz, there are more women going into the workplace in employment generally. Has that had an effect on inequality trends generally?

Liz McKeown: There is probably not a simple answer to that, because it would depend on what inequality trends you were looking at, over what time period, what contribution that was making, whether you were looking at the individual level of the woman or the household, and what have you. We have to understand exactly the specifics there.

Q68            Julie Marson: I do not know if Torsten wanted to say anything, I am just conscious to leave a bit of time for my colleague.

Chair: Be really brief, Torsten.

Torsten Bell: There are some studies that look into that. The first thing to say about having more women in work is that, in our view, that is a very good thing. That does other things, but it is really important. We should be aiming for more gender equality.

We are in favour of much more aggressive action that makes men actually take some paternity leave at the beginning of their parenting. That is the only way to deal with some of these longerlasting problems, because we have made progress on the gender wage gap before kids. Kids are the problem, as it were. Unless you do something about that, we are not going to be dealing with the other problems.

In terms of its effect on the economics and on inequality, it could go either way, in theory. In practice, the increase in both people in a couple being likely to work is a result of assortative mating, i.e. people are more likely to couple up with someone who is similar to them. They are more likely to have met in university. They are not marrying their secretaries anymore, if you see my point.

The effect has been to push inequality up slightly from that perspective. There are arguments about the different degrees of it. What we do not discuss enough is how it is having an effect on hours worked. What we are seeing is spreading: rich families are working lots of hours and poor families are not working many hours, with both working part time, for instance.

Q69            Harriett Baldwin: Can I start with Liz? We received written evidence from the TUC that said that, during the pandemic, six out of 10 people who died from coronavirus were disabled. I just wondered whether that unbelievably startling and shocking data point is something that the ONS has been able to evidence.

Liz McKeown: I have not seen the TUC analysis, but we have looked at mortality by different characteristics. That has been a really important part of the work that we have done during the pandemic. We do see greater risk of death for people who are disabled compared to those who are not. Looking at a particular between January and November 2020, the risk of death involving Covid was 3.1 times greater for more disabled men and 1.9 times greater for less disabled men compared with nondisabled men. For women, the risk of death was three times greater for more disabled women and two times greater for less disabled women compared with nondisabled women.

You are seeing some really very striking figures there. We did some analysis to understand how much of that was driven by disability and how much was driven by the different characteristics of disabled people compared to nondisabled people. I can provide further details on that analysis. It is quite indepth, but, yes, we did see that increased mortality risk.

Q70            Harriett Baldwin: It could plausibly be as high as six out of 10 people.

Liz McKeown: I would not like to say what that translated through to in the numbers. I would rather take that away and come back to you with a view on that, but we can clearly see the increased risk.

Q71            Harriett Baldwin: Do any other panellists have any data on that?

Torsten Bell: No, we have only focused on the labour market effects on disabled people.

Q72            Harriett Baldwin: What percentage of the overall population would normally be described or selfdescribed as disabled?

Torsten Bell: It is in the 15% ballpark, but that is very easy for us to write to you and confirm.

Q73            Harriett Baldwin: It is clear that a lot of the shocks we have had, not only on the health side but also the economic impacts—we are focusing on the recovery herehave exacerbated existing inequalities.

Specifically focusing on disability now, as we look forward, in terms of enabling a more equal recovery, I wonder whether our panel think that, for example, the move to more working from home and the move to different sectors in terms of the recovery are likely to help or hinder the disability pay gap, which everyone would acknowledge exists. Liz, I would like to start you with you and then bring in the other panellists.

Liz McKeown: From the labour market perspective, prior to the pandemic we saw the employment rate between people with disabilities and those without narrowing.

Q74            Harriett Baldwin: It was still very wide.

Liz McKeown: Yes, but that was the trend. Over the pandemic period, the impact on employment has been similar for both groups. While it suggests that, in labour market terms, those with disabilities have not been disproportionately impacted

Chair: I am sorry to interrupt. The Division has just been called. I do not know how many questions you had. I do not want to put any pressure on you at all, but, if you feel that you are satisfied the progress you have made so far in your section, that is fine. If not, I will suspend the Committee and we will come back.

Harriett Baldwin: Given that our panellists have both mentioned that they would follow up with written evidence, that would be really helpful, if there were to be any further evidence that you wanted to give on this particular question. I was going to focus on disabilities and also different ethnic groups, if there are any observations that you want to specifically share with the Committee in terms of the inquiry that we are doing here. Then I am happy to leave my questions.

Chair: That is fine. If you could take that as a request to write on those matters, that would be very helpful. I am sorry; unfortunately, we have literally run into the Division. Thank you so much for joining us. It has been very helpful.