 Welcome to the 28th meeting of the Economy, Jobs and Fair Work Committee. I would ask everyone present to turn electrical devices to silence so as not to interfere with proceedings. I have apologies from Gil Paterson, committee member, and I would like to welcome Alec Neal, who is here as substitute for Gordon MacDonald and will be with us for the next few sessions, so I would invite Alec Neal to make a declaration of any relevant interest. I'm a part-time adviser to ethics energy limited, and I'm also a co-convener of the cross-party group and Brexit in the Scottish Parliament. Thank you very much. Item 2 is a decision by the committee to take items 4 and 5 in private. Are we agreed on that? We turn to our economic data inquiry, and I would like to welcome our witnesses today. We have today Sir Charles Beane, a member of the Budget Responsibility Committee, I think, in the Office for Budget Responsibility. Ed Humpherson, who is director general for regulation of the UK Statistics Authority, and Jonathan Athau, who is deputy national statistician and director general for economic statistics in the Office for National Statistics. Welcome to all three of you this morning. If I might start with a fairly general question, and that is what you think of the current provision of economic statistics in Scotland, what you might view as gaps in the coverage in those statistics, and what specific recommendations you would have for improving. As we go through the questions, please don't feel you have to answer each and every question, but do come in and indicate if you want to come in by simply raising your hand. I don't know who would like to kick off with that particular question for myself, or questions for myself. I'm actually going to throw the ball to Jonathan on this for quite a good reason, because in my review, which came out well over a year ago and now, it flagged that one of the areas where more work was needed was regional statistics in general, but it recognised the problems of doing that. In particular, if you want lots of detailed regional statistics, you obviously need lots of information, and the finer the degree of disaggregation that you want to look at, the more information you have to collect. I had basically seen that really the only way of unlocking this was access to administrative data, which, of course, the digital economy bill has facilitated. I do know that the Office for National Statistics has actually been moving forward in plugging some of the gaps that I had flagged in my review. It's probably better for Jonathan to talk about the work that they are actually doing at the moment and things that they have planned in terms of the regional statistical estate, and, obviously, particularly apropos Scotland. I might start by saying that we are on quite a journey here. The way in which devolution is changing in Scotland has changed needs and user needs of where we start from, in terms of thinking about our statistics. Following Sir Charles's review, we have now put an added impetus on to making certain that there is additional geographical detail available. That applies to Scotland, but it also applies to Wales, Northern Ireland and the English regions as well. It is something where we have done a few things differently recently, so we published public finance statistics for all the countries and regions of the UK building on some work that is done in the Scottish Government through JAERS. Things can be done and things are moving. Next month, we plan to publish something called a balanced measure of gross value added, so gross value added is akin to GDP. Previously, we have published statistics for Scotland, but there have been two different measures using the income and production approach. We are going to produce a single measure to give users in Scotland a much clearer picture of economic activity in Scotland. There are lots of plans in place to improve statistics, but that comes from user needs. That will come from what the Scottish Government wants. It will also come from what other users in Scotland are looking for. We are thinking quite widely about that. We are publishing Thursday this week a study looking at how practical it would be to publish a different inflation estimate for the countries and regions of the UK. That has always been something that we have thought has been quite difficult because of the way that prices are collected to inform inflation statistics, but that is exactly something that we are looking forward to think about, whether we can provide that sort of information because that has been a gap that has been in the past. The answer is that we want to be as responsive as we can be to user needs. There are real challenges, as the child said. In some cases, the way that samples are collected for statistical purposes, the surveys that are collected, means that it is very difficult to go to very fine levels of geography. Often people want not just Scotland, but particular geographies within Scotland. Obviously, there are city deals emerging in Scotland, so people want to understand what is going on in their local area, but that is always going to be very difficult when collecting data from surveys. One of the things that we are now trying to do is unlock the potential of administrative data. One of our first data sets is to use VAT data from HM Revenue and Customs that allows us to understand what is going on with economic activity in a particular area. That should allow us to have a better understanding of what is going on. We are only starting to explore that data, but it is that data that underpins our balanced gross value-added statistics that I talked about earlier. This is an evolving area. There have been some gaps, but we are trying to fill them. As I said, coming back to what I was saying at the beginning, we are guided by user need. If the committee comes out with particular areas or particular gaps that it has heard from users, that will be useful to us in informing our work programme and our conversations with the Scottish Government about how we work together to make certain that there are good economic statistics for Scotland. My perspective is of the person who heads up the Office for Statistics Regulation. As such, we oversee statistics produced in any part of the United Kingdom. Certainly produced by Jonathan and his teams in ONS, but also by the Scottish Government, Whitehall Departments and the Welsh Government, the Northern Ireland Administration. That gives us a really good capacity to look across the whole provision of statistics in the round and form an overview. From that perspective, I would say that Scottish economic statistics produced by the Scottish Government were in a good place, in two senses. Firstly, it seems to me that there is a very clear framework for the development of the statistics and the production of the statistics, starting with national accounts, with some good sectoral information on different sectors of the Scottish economy, and with some good labour market information as well. I think that there is a clear framework. One of the things about a clear framework is that it really helps you much more clearly to see where the gaps are. It is harder to see where the gaps are if you get a fairly random collection of data releases. With a clear framework, you can see where something is stronger and weaker. There are some weaknesses. There are clearly some weaknesses. That brings me on to my second point about why the Scottish Government is in a relatively good place, which is that there is an appetite for improvement for addressing the gaps and a track record of having done so. You heard a lot from your witnesses in previous sessions about the improvements that have been made over time to Scottish economic statistics. Those gaps are significant. Jonathan has mentioned prices. Those data until recently have not been available in Scotland. There are questions about trade information, exports and, in particular, imports. There are also questions about the timeliness of regional economic statistics. I think that one set of evidence that you received stood out for me, which was the Scottish Fiscal Commission talking about labour market information, income and earnings and how more timely and more comprehensive information on income and earnings would really help them at forecasting the economy and forecasting income tax receipts. That strikes me as being a very salient issue that would be very sensible for the Scottish Government in partnership with ONS to address. We see a good base, but there is clearly an appetite for improvement and some things need to improve. John Mason You touched on a lot of issues between you there that I think we are going to try and explore a little bit further. The Digital Economy Act has been mentioned and the fact, I think, that we are moving towards a bit more administrative data rather than surveys. How do you see the UK picture at the moment? Is the Digital Economy Act going to fix everything or are there other things that we are needing to move on? How does the UK compare to other countries? We hear a lot of good stuff, but it tends to be often about small countries such as Estonia, Denmark and the Netherlands. Maybe that is just not possible at the UK level, so I am interested to see where you all see that the UK is going on this. John Mason If I could just start on that, it is not just small countries that have made effective use of administrative data. Canada, who is regarded as one of the best national statistical institutes, rely very heavily on administrative data. Basically, whenever they want to run a new survey or something, they have to make the case that they cannot get the information from existing data sources. It is very notable that the Canadian statistical authority is a much more innovative, agile institution as a result. It is regarded as one of the best employers in Canada, so it is one of the best places for economists to work and things like that. I do not think that you should think of administrative data as being just a preserve of small countries. I think that there is a lot of potential there, but equally it is not going to solve everything. The key thing about administrative data and private sector big data is that the information is not generally collected with statistical use in mind. It is collected as a by-product of collecting taxes or if it is in the private sector supermarkets pricing their products and registering what is being sold and so forth. Then it is a question of statisticians levering off that, so you need to use judgment and expertise in the way that information is exploited. It needs to be used often with surveys, where you need surveys to fill gaps and so forth. I certainly took the view in my review that, while unlocking administrative data was certainly helpful and an important step forward, it is not a silver bullet that is going to solve all the UK or Scotland's statistical problems. Just as far as the act itself goes, it has ended up going further than I thought was possible. In my review, I focused particularly on facilitating ONS access to administrative data held elsewhere in government, most obviously the tax authorities. However, it is not solely the tax authorities, because obviously lots of other departments have similar information that may be potentially of use. I did not really wade too much into the access to private sector data question. In fact, the act, as implemented, gives the ONS rather more scope for using private sector data than anticipated, but most importantly, because this is the first step, it removes the obstacles that were there to effective use of administrative data across Whitehall ministries and so forth. There need to be protocols put in place, which I think that the ONS is presently consulting on, but I am pretty happy personally with where the digital economy act came out. Jonathan may want to expand on some details of how he anticipates his operation, but we are in a much better place now than we were. Were we comparable? Were we maybe not comparable with Canada, but were we comparable with Germany, the States? Potentially, we could be comparable with Canada. That is what we should be aspiring to, and not just comparable to overtaking them. I think that underpinning your question was, were we keeping up with leading practice? On administrative data, we were not in the first group of countries who were using administrative data, but now we have the digital economy act, and we were able to learn from how other countries had done it. Other countries had focused primarily on Government-held data. We could think that there is advantage in having powers over private sector data. Some of you may have seen, and it was reported in some of the newspapers this morning, we have been looking at mobile phone data to understand commuting patterns, and how people travel to work from it. We know that mobile phone data could be a rich source of understanding of what is going on in the economy. We know that the private sector is also very important. We have also been able to learn from other countries that have gone first in terms of the techniques that they have used. From December this year, we will be using VAT data to put together our national accounts, to put together measures of GDP. We have worked closely with the Dutch statistical office because they have done that for a number of years. We are able to learn how they used it. One of the points that Sir Charles makes that I think is very relevant is that this data is not primarily for statistical purposes, and you need to invest time in understanding it. You might have to use different methods, different statistical techniques to understand it. You might find that it fits some parts of the economy very well, so we found some areas, some sectors of the economy, where the VAT data describes what is going on very well and other areas where it does not. That is sometimes understanding the way that it is collected, so VAT data for some companies will send in one return covering a large number of businesses. It is sometimes difficult to apportion how much economic activity is happening in a particular company or a particular local organisation on the unit. Just specifically on that point, they would not be splitting it between Scotland and Wales. Exactly. You have to understand the nature of how that data is submitted, whether you want to apportion it or whether that is an area in which you still want to use a survey. For some of those complex companies, you might want to ask them a survey because it is better than the administrative data. If I can just press you on that particular point, the Scottish Government does not have access to VAT returns and so on, and even if it did, it would be possibly at a UK level. Is there something that we can learn? Is there something that we can take forward? Should we look to you to get a bit more detail or should we be doing something ourselves? We want to use that data to understand industry breakdown and geographical breakdown. For some companies, that will work very well. If you are, all your activity is based within a particular area, the VAT data will work very well for that. It is just about understanding where it works and where it does not. That is something that we will want to talk to the Scottish Government to understand what the Scottish Government data needs are. We can work through how we meet those needs. Is it administrative data? Is it survey data? Is it a blend of the two? How do we go about doing that? One of the things that we want to start doing as we go forward is making the underlying data available. One of the revolutions in statistics in the last 10 or 20 years is that more of the underlying data is now made available for anybody, whether that is an academic researcher or a Government researcher, to access. O&S data, although that is mainly survey data at the moment, is available to researchers to interrogate. We would want to make certain that, as we get these rich data sources, those data sources are also available for people to then do their own analysis, because we cannot cover every particular cut of the data that people want. We want to make that data available for people to be able to do their own analysis, but we also want to work very closely with Scottish Government and other users in Scotland to understand what their needs are. We have received quite a lot of evidence, both oral and written, about the poor response rate to key surveys that are issued by the Scottish Government. There is a global connection survey that looks at export data, where the response rate is pretty low. One of the things that we are examining is what can be done about this and is the solution to give to the Scottish Government the powers that O&S has got, for example, or the Northern Ireland statistical agency has got, to compel businesses, for example, to respond to surveys. I wonder whether you have a view on that. I start on that. With business surveys, we have under the statistics and registration services act. We have the power to compel businesses. In general, response rates from businesses given that basis are quite strong. In some cases, some of our surveys have a particular Scottish, what we call a boost to the sample. Of course, we can use our powers there to make certain that boost that the Scottish Government wants is collected effectively. Within the business surveys, I do not see that there is a particular issue. The issue for businesses is one of burden. We know that businesses would rather be getting on running their own business, rather than filling in forms for us, but that is where, for example, using administrative data could be very helpful that we can reduce the burdens there. For individuals, there is a real challenge about response rates. We know that fewer and fewer people pick up their landline telephones. When they do, we all know whether it is somebody trying to talk to us about PPI or trying to sell us something else. There is a large degree of scepticism when we are trying to collect data over the phone or even sometimes face-to-face. With individuals, it is a challenging environment. We are seeing response rates for our individual surveys falling. That is a concern because falling response rates could lead to some biases in the surveys if you are not very careful at understanding what is going on. That is a particular challenge. We are finding that in particular on our labour force survey, which underpins all of our employment statistics, that fewer and fewer people are willing to participate in that. What we are trying to do is get under the surface of that. We can shorten the forms, shorten the questions that we ask, because sometimes it is the fact that you have to be on the phone for 30 minutes or so and put people off. Are there behavioural insights that we can use, nudge techniques that we can use? Are there ways that we could provide incentives for people to respond? We are trying a mix of those different things, but it is a challenging environment. Many countries around the world are finding that people are less willing to participate. Again, that is something where we can use administrative data to take some of the questions out. We will ask people what their income is, but if you are an employee through the PAYE system, that information should be available. Are there ways that we can use administrative data to reduce the burdens and prop up response rates? It is a real challenge around individuals. As I said, businesses are less concerned about burdens on them. On the specific point about global connection survey, I will not give a view on the general power to compel, but if you look at the global connection survey, you are right that the response rate is relatively low. It is something like 1,700 responses out of 5,500 surveys sent out, something of that order. Would you want to compel if you had the power? The question there is balancing the cost, which would be the burden—the respondent burden—with the benefit and the benefit that you would have to think about. How much more information would you get from a more complete survey? There are statistical approaches to saying what is going on with the non-respondence bias. It may well be that 1,500 is giving you a sufficiently reliable picture because there is no difference between the respondents and the non-respondents. It could also be that the non-respondents have a very different structure or a different ownership and therefore you are missing quite a lot of information from those missing responses. I think that that is the way to think about it. What would you get from a greater response rate and does that merit the costs of compulsion, where you would have the power to compel? One of the things that might be worth injecting into this, which connects a little bit with Jonathan's comment about nudge techniques, is that you can give an incentive to the person or surveying to complete the survey or provide the information so that he or she gets something in return from doing that, whether that is financial or there is something about what the survey is used for, or that that information can be provided back to them, which helps them in their business or whatever. That is another way in to do it. At the moment, very often surveys, it is all burden on the person filling in the form and their time and so forth, but if they are getting something in return, then they may be more likely to participate. I mean, it seems to me, and maybe this is just me being a bit too naive and altruistic, but it seems to me that in order to inform public policy, knowing what trade patterns are and where there are gaps and where there could be boosts, could benefit those individual enterprises. So, you know, that seems to me to be an area worth exploring. The other thing that crossed my mind on my less altruistic side of my brain was what remedies do you have if people do not co-operate? I do not know the powers, but we certainly have the power to find businesses if they do not comply. That is obviously something that we do not want to do, and just coming back to your point, I very much agree. There is a challenge sometimes as a statistical office. You will talk about, well, this is very important for the point of view of the survey, but then actually you need to tell the wider picture as to why these surveys are very important, and we find that also is very true on the individual survey side. We explain to people why, if we are coming around to understand about your wealth assets, your pension savings, about how that informs government pensions policy. So, our best field researchers are the ones who make those connections and explain why. I could not agree more that these data feed into very important decisions and decisions that affect businesses, individuals in Scotland, and it is in some ways a public benefit through the statistics that they help to provide. We have heard from witnesses that statistics on non-market activities and well-being are not as well covered. Is that just because there is a lack of demand for them, because the OECD certainly talked about other countries using time-use surveys, and I wonder whether the UNS is considering anything similar? So, maybe I should start by saying, we certainly see there is a wider need to go, the phrase we use is sort of beyond GDP. Now, obviously, GDP is very useful, it informs a lot of forecasting around the economy, around tax collection, and revenue generation more widely. So, it's really very important as a concept, but we have used time-use surveys to look at non-market production. We published some results on that last year. Again, that does show some interesting trends. Actually, non-market production has grown slightly quicker than GDP in the recent past, so that is out there. It's something we'd like to think about how we can make it more sophisticated, but that is quite burdensome when you're asking individuals for how did you spend your time today. You can see quite quickly that that becomes quite burdensome, but it does give you insight into what's happening with patterns of childcare or caring for the elderly, how much volunteering goes on and how much that's worth. So, there is quite a lot in that space. We've published our well-being statistics, we've published the next batch that came out this morning. That's something where we are trying to make certain that we are providing a wider picture. An area where the UK leads is on environmental accounts, understanding systematically the benefits that accrue from biodiversity, from different sorts of environmental amenity. Those are areas that we are trying to pursue, because they are very important to people's quality of life, and people want to know what is happening with, for example, our forests or peatland or things like that. Those are areas that we are trying to do more. Historically, I think that GDP, because it is so well developed, because it is so well resourced, is in some ways decades ahead in terms of development, but that does mean that those other interesting areas can catch up quite quickly. We try to make certain that we cover them. We obviously have a limited amount of resource and have to balance that quite carefully, but the issues that you raise are exactly ones that are very much prominent in our thinking. I think that it's important to distinguish two separate strands in your question. One is non-market activity, and the other is the broader question of wellbeing and those other indicators. On the first of those, one of the issues that I discussed in my review was the fact that there might be shifts across the production boundary taking place. Conventionally, statisticians have said that they will measure what takes place in the market economy and not the stuff that takes place in the non-market economy. There's a lot of economic activity, work that's done at home, charitable activities and so forth. With some exception, imputed rent of owner-occupiers is a classic example of something that is not within the conventional production boundary, but we measure in the national accounts. The reason for that is that you don't want to distort the conventional market measures if there's a shift between owner-occupying rented accommodation. Activities that are sufficiently big and might change in importance, statisticians would think that there's a case for measuring them. At the current juncture, it's quite plausible to say that there's quite a lot of activities that previously would have been done by intermediaries in the market economy. I always give the example of travel agents. In the old days, you'd go down when you want to book a holiday, spend an hour talking with a travel agent about where to go, then he'd get on the phone and book the flight to you, book the hotel, you'd go and buy a guidebook. What you do these days, you do the research online, you book direct with the carrier, maybe stay Airbnb rather than a conventional hotel, you use Google Maps to get around. A lot of that activity has been taken out to the conventional market economy. Potentially, the market economy statistics like GDP are being distorted. The solution is not necessarily to say, well, we have to expand the measure to include all these other things because they may be difficult to measure continuously, but as a minimum, you want to know how big the distortions might be, which is whether these time use surveys might come in. I suspect that we will need to do more of that going forward, but there may be clever ways of getting information on that. Obviously, internet usage is something that a lot of data is collected by the digital companies. The broader issue of wellbeing, of which there has been a lot of interest recently, I should say from my perspective, talking just as an academic economist, I am sceptical of single measures of wellbeing, gross national happiness, things like that. I certainly do not think that GDP is a sufficient statistic for welfare or anything like that, and economists will always tell you that, but I do not think that the solution is to get a load of indicators together, throw them in the pot, put some arbitrary weighting system in and say, here is my better measure. GDP works as a concept because it can use prices to weight things together and prices are a measure of marginal utility. As soon as you start incorporating other things, it is not clear what the weighting should be, so I think that a more fruitful way forward is something like a scorecard approach, where you decide what indicators you feel you need to look at, and instead of trying to add them all up, you have your 20 or so indicators that you might want to look at, and you look at the detail and discuss which ones you, as a policymaker, think are the most important. I very much agree with what Sir Charles has said about the distinction between GDP as this, in effect, internationally comparable and consistent measure, and then something that might be more of a dashboard or a series of indicators that responds to different policy imperatives and policy drivers. One of the really interesting things that you, as a committee, have been hearing from your witnesses has been the kind of fault line between what needs to be done to replicate at the Scotland level the best possible range of economic statistics that are available at the UK level, and that is what lies behind a lot of the gaps that I mentioned in my introduction, is that sort of that replication. That is important, it is very useful and very much needed. The fault line is between that and then thinking, what are the things that we actually care about in the context of Scotland as policy makers, as people who scrutinise Government performance? There, I think, you are not so much in a space of replication, you are in the space of innovation, of thinking what are the things that really matter. You have talked a lot about the four eyes, I think they are a particular emphasis in Scotland's economic strategy, and I think developing a way of getting a really good handle on those is something that could be quite innovative. One other thing that I would mention, Jonathan talked about personal wellbeing measures which were published this morning by ONS, I think they are relatively underappreciated as a resource in this space actually. They are often spoken about slightly dismissively as being a kind of a happiness measure. I think that they now is a growing time series of personal wellbeing over time. It is quite possible to drill down to quite local levels because the dataset is large enough, and it is quite possible to correlate the personal wellbeing responses that people give with other aspects of their life. I think that that is quite a rich policy area to think about wellbeing, which has tended to be slightly marginalised. I would encourage some creative thought there as well. I am curious to know if we are happier or not as a result of the survey data. Just one supplementary question, because obviously you are starting to collect that wealth of detail. What is the size of the Scottish sample? Would you advise us to collect those ourselves or boost the sample? I do not know off the top of my head, but I know that it is a representative enough sample for Scotland, but I can come back to you with the exact number. The way that we tend to do it is by tagging some of those questions on to existing surveys. Often there are a small number of questions and we will have gone out to talk to somebody about if they are working and their job, those sorts of things, their pensions or something like that, and then we will tag on the question at the end. It is often quite easy for us to add a couple of questions to the end of a survey. It does not really add much to respond to the burden, but it gives us this much richer source of data. The whole person perspective, because you can relate how an answer on a personal wellbeing question relates to what they have said in other parts of the survey about their employment status, or their pension status, or their health, or whatever it might be. That is why it is quite powerful. For example, one of the bits of analysis that has been done has been looking at how happiness varies according to people's sexual identity and issues there. It does give you a very rich data set that goes much beyond economic statistics to understand wider social phenomena. A follow-up from Alex Neil before we move to a question from Ash Denham. Can I just go back to what you said about VAT receipts? Clearly, the Scottish Fiscal Framework Agreement between the UK Government and the Scottish Government has a notional figure in there for VAT receipts. From what you said, it sounds as though the calculation of VAT receipts allocated to Scotland may not be a reliable figure. The key thing is to understand that there is a distinction between how VAT is collected, collected from companies, but the incidence is borne by individuals. I do not know exactly how the methodology works, but if you are interested in where the receipts are generated from, you are more interested in where people are spending rather than in which company collected it. You can develop a methodology to look at that. You are answering a different question. It goes back to the point. Understand how the data is collected and whether the VAT collection data probably does not help you to understand where the burden of that VAT is paid from. What is the methodology used for calculating the treasurer's calculation of that allocated to Scotland? I do not know off the top of my head how that works. Could you come back to us? We could come back to us, but I am certain that the Scottish Government would have been also involved in that work as well. The problem is that the Scottish Government does not have any statistics that rely entirely on the UK HMRC and the like to provide the statistics. The other reason why this is important is that there is a political controversy at the moment about VAT on police and fire services. It is interesting to know whether the methodology can state or not state explicitly if the VAT receipts from the police and fire brigade are included in the notional figure for VAT. Clearly, if they are already included, the situation would be neutral. If they are not included, then there is obviously a deficit. That would be useful if you were able to answer that question. Can I also go on to—obviously, you referred to JAERS—and, obviously, there is a lot of very good work done North and South of the border on JAERS down the years. The original version, I think, was published in Roy Jenkins, with the Chancellor of the Exchequer, and we back in the late 60s. As well as VAT, clearly there are big numbers in there for corporation tax and capital gains tax, for example. How reliable are they? Often, when you are getting into an apportionment, you need to make some assumptions. For corporation tax, for example, we know from other sources how the workforce is distributed across Scotland and other parts of the UK. Sometimes you will use employment as a proxy for where the economic activity is taking place. Sometimes it is using proxies rather than actual understanding of where the tax arrives. Obviously, if there is a company that has a single office and only operates in Scotland, that is quite easy to understand where that is. It is more complex when you have the supermarkets, for example. How do you allocate those? There, we would look for things like employment to allocate that. Again, that may not be a perfect approach, but, often, because the way that the tax collection system works, it does not allow you from the tax collection to identify where the economic activity arose, you have to use proxies to do some apportionment. If we go back to Canada, because it says in the briefing here that, in Canada, most of the statistics I have gathered are top-up rather than bottom-up rather than top-down, in Canada, by collecting the statistics, bottom-up rather than top-down, is that why they are able to produce more accurate figures at a provincial and regional level? I do not know enough of the details of how Canada works, but I know certainly, for example, they apportion their VAT system and, again, they have to use a whole series of proxies for that. In theory, you can start asking businesses to apportion their activity between different geographical areas, but that can be quite burdensome. As a business yourself, how do you allocate overheads? It is always going to be a challenge. How do you do that? Even if you start asking more granular data of businesses, that might not be any more reliable than using other data to apportion the economic activity. What is likely to be the margin of error, for example, in co-operation-type estimates for seeds? I would not know off the top of my head, but I can give you a note as to explain how that will be very helpful. My question is specifically directed at the ONS. We have a situation where levels of devolution across the UK are somewhat variable in terms of statistics and so on. The ONS focuses particularly on developing statistics for both the UK as a whole, but also on regional statistics. To what extent are the unique or specific needs of the Scottish Government or Scotland as a whole factored into the primary objectives of the ONS? Going back to my opening remarks, everything that we do has to be user-focused. What are people needing those statistics for and decisions that need to be made? In that sense, we are very, very alive to the needs of Scotland, but as I said, Wales, Northern Ireland and the English regions with devolution within England. That is a real focus for us. We have a track record of working with the Scottish Government to innovate, to do things differently, to boost sample sizes in Scotland to provide more detail where that is needed. From our point of view, there is a real willingness to work together. We have a regular meeting called the Interadministration Committee where we sit down and we compare priorities, but that is not just economic statistics. That includes wider statistics population and those sorts of things. We have a particular group that meets on economic statistics between the ONS Scotland, Wales and Northern Ireland on a quarterly basis to look at common issues, to look at what needs are. I think that there is a lot that we do to try to make certain that we incorporate all the different needs, whether that is a UK-wide need or whether that is a Scottish need. As I said, it times some things that are easier to deliver than others. Some things require extra resource or new data collections to do, but we try as far as possible to meet all the user needs. I think that I would hope that we have a good reputation with the Scottish Government for working constructively with them. When we assess statistics produced by ONS, which are at the UK level, we assess them against the code of practice to determine whether they are meeting the high standards to be a national statistic. One of the things that we look at is the extent to which they are serving user needs beyond the UK level. We recently published an assessment of regional gross value added and we looked a lot at the extent to which the questions that those statistics were addressing spoke to the interests of policy makers in Scotland and in the regions of England, Wales and Northern Ireland. It is a thing that, in a sense, my job is to prod or nudge various degrees of politeness to firmness the ONS into making change. That is an area that we do make those interventions on. That was what my next question was coming on to, the process for developing those objectives. You have touched on that with the Inter-Administration Committee. You know that Mr Humphreys said that there is possibly an element of encouraging the ONS to develop those processes. If the Scottish Government, for instance, came to one of these committee meetings and said that we really want to go in a different direction, we want data on X, what would the turnaround timeframe be in terms of developing that new statistics? I would be slightly unhelpful to depend on what it is. As I said, inflation statistics have been a long-standing issue where people wanted to know more. We were lucky in the sense that we have now brought in a wider range of academics to work with us, so we can now get them to think about how realistic that is and what we would need to do. In some cases we can respond very quickly because if the data is already sitting there and it just needs to be cut in a different way, that can often be done quite straightforwardly. If it requires a whole new data collection, we would just sit down with the Scottish Government and work out what the best way of doing that is. That might be a combination. Scotland runs some of its own commission surveys itself. It could be something that we do, it could be something that we do where we take one of our existing surveys and modify it slightly in Scotland. All those options would be open and we would just think about what the best way of meeting those needs are. Sometimes there are very difficult conceptual issues. Understanding exports or imports to or from Scotland is conceptually challenging. It is not just in Scotland, but Wales is also very interested in that, but that is quite challenging. Sometimes it is not just the data collection, but it is also the conceptual issues of what does an export from Scotland mean, for example. It can be quite challenging to measure those things. I think that we would just be very open-minded about that. There are lots of different ways of answering those questions. Can I just inject a suggestion? At the moment, key users, such as the Bank of England and the Treasury, have an annual process of interaction with the ONS, on which we basically each year would say what worked well, what hadn't worked well, but also signals priorities for future development, what we would like to see. When I was at the banquet, we would often say that we would like to see completion of the flow of funds, for instance. There is a question about to what extent the Scottish Government is consulted formally in its role as a user of statistics and whether it might usefully be brought into that process. It is simply a mechanism whereby, on a regular basis, it could flag up areas that it thought would be developed. Following on from section 20 of the Statistics and Registration Service Act, it provides that the board may produce and publish statistics relating to any matter relating to the United Kingdom or any part of it. Can we take it from that that the UK Statistics Authority has got unfettered powers to produce any statistics for any part of the UK on any matter? That is just what that says. In certain areas, such as the collection and publication of population statistics rests with the National Register for Scotland. In some cases, there are specific elements of devolution that have meant that, for example, we are the census provider in England and Wales, and Nisra is the provider in Northern Ireland and NRS here in Scotland. Broadly, yes, but there are some areas where there are long-standing constitutional differences, but we do not get into those particular areas. In section 23, it says that the board may not, without the consent of the Scottish Ministers, produce and publish Scottish Devolved Statistics. I presume that the board has never tried to publish devolved statistics, and the Scottish Government has said no. No, indeed. That would be, for example, population statistics that I was just talking about. It seems that there are no statutory constraints to getting better statistics that are relevant to users, particularly in Scotland. Last week, we heard from the Scottish Fiscal Commission who perhaps have a need that is of a different order of importance here, because they are involved in forecasting. On the basis of those forecasts, we set budgets that affect the amount of tax people pay, the amount of economic activity that we anticipate to be generated in the economy. Given that the current arrangements between the statistics authority and the Scottish Government are via protocols and memorandum of understandings and joint meetings that take place on a regular basis, I understand that those arrangements work pretty well. Is there a need to formalise that along the lines that Charles Beane has just talked about to make sure that that becomes part of a much more formalised process whereby the needs of the Scottish Fiscal Commission can be flagged up pretty early in a process and once it takes on its powers, everything is in place to allow it to do the job to the best of its abilities? I think that there could well be. I think that one of the things that we have been thinking internally is that through the UK Statistics Authority board, we are accountable to the Scottish Parliament, the Welsh Assembly, Northern Ireland Assembly and the UK Parliament. We have been thinking about how we make that real. We obviously lay our annual report and accounts before the Scottish Parliament, but could we do something more systematic that would talk about what we had done to meet Scottish user needs and use that to elicit further feedback? I am very open-minded about how we might strengthen that relationship to make certain that all needs are articulated early. We go out and engage all users. We have something called NACOMIC statistics and analysis strategy, which sets out our priorities for the year. We go out and consult on that and say is that people's needs, but certainly if there is more we could do that would be relevant to Scotland and particularly the Scottish Parliament, I think that we would be very open-minded about that. I think that there is something in what Sir Charles has suggested here. I think that the Scottish Fiscal Commission is a really good exemplar of it. Here you have a body with some really important new responsibilities that has identified a specific need or lack of completeness in the data that is available for a really fundamentally important thing that they spoke about. The time lag for their access to income and earnings data and thinking about whether there are more timely ways that they could get insights into income and earnings in Scotland. Of course, that resonated with me because the Office for Statistics Regulation has, for a couple of years now, really been prompting on the question of more comprehensive income and earnings statistics and more timely ones. The general point that here you have in the Fiscal Commission an important user with new responsibilities would be exactly the sort of player who could feed into some kind of forum to say, well, these are our needs. What are the best ways to meet them? On that point on the practicalities of doing this, we have some figures about the money that the Scottish Government spends on boosting the various surveys to get better results. There are modest sums of money in the bigger scheme of things, but can I take it that those kind of arrangements are ad hoc arrangements? The Government says that it wants to boost a survey, talk to ONS, you say, well, this wouldn't be very easy, budgets are tight this year, they say, well, we're happy to chip in something to help, and then you have a negotiation. Is that the kind of way it works or is there some formal protocol to what the Scottish Government will pay for and what it won't pay for, what it expects you to do? I mean, essentially, we would set a requirement to meet the particular needs, so that might be a certain sample size. If Scotland then wanted an additional sample size, we would expect that on a cost recovery basis, unless there was something very trivial. If it was a very simple thing to do, we wouldn't ask for cost recovery on that, but more generally, it's where it differs from the UK approach that we would expect. That's how it has worked, that we ask for cost recovery on any additional costs from boosting those samples. We think that that seems to work well, gives Scotland the statistics it wants. It doesn't then mean that we have to trade off anything else within the ONS. That then becomes a question for Scotland to think about whether that is worth the additional cost of boosting those samples. Gillian Martin You've largely covered quite a lot of what I was going to ask about inclusive growth and regionality, so I'm going to ask you something else and forgive me if it sounds like I'm expecting you to have watched every single second of an inquiry so far. Tell me if I am being awful at it. Our inquiry has been going on for a number of weeks. Is there anything that you've heard from evidence and prepared to come today that struck you as being something that you didn't already appreciate or know that might have given you pause for thought on how your existing bodies can better serve Scotland's interests? Maybe not with what we've just talked about, Landie Whiteman's question about us paying for more, but changing slightly how you operate when you're doing the existing surveys? I think that picking up on the Scottish Fiscal Commission is something that is rapidly developing an interesting and challenging new function that needs to be done. I think that there is something there that stood out for me as a particular issue. The other thing that was interesting here was that we are at a very interesting stage in the sense that there are huge numbers of new requests coming through and I could see lots of different elements that people were asking for. I was surprised by the breadth of it and I think that simply as devolution widens and deepens, so I think those needs will then become a bit more focused as new frameworks have developed as the Scottish Government gets grips with new powers. I think that those will then become a bit sharper focus. The breadth of what you've heard covers virtually everything in my responsibility in the ONS and a lot wider. It really was a real tour of all the issues on people's minds. Which of those have become the most prominent in terms of really crunchy statistical needs? I think that's the next stage that I'm interested in seeing how we narrow down that very wide field into some priorities that we can then start to work through. From my point of view, and I'm not sure if this is going to answer your question, but I'll say what my reaction was to looking at the previous sessions. There's one thing which slightly surprised me and one thing which quite excited me. The thing which surprised me was I think you had a couple of witnesses who talked about the need for a sort of stronger ecosystem of research and analysis outside government in Scotland. That surprised me because, from my perspective, I see quite a strong range of organisations both in academia, there are some good think tanks, I think Fraser of Allander is an absolutely superb institute which does great work. So I was quite surprised at some of your witnesses. I wasn't quite sure what their benchmark was, I think there's a really good basis. The thing which excited me builds on and that is that you had a few witnesses who were talking about not so much the statistics and data produced by government, but the statistics and data produced by a whole range of other actors, enterprise bodies, research bodies and so on, and thinking well is there some way of sort of linking those up into a consistent framework because of course then it'll be more usable. I think that's really exciting and it really maps on very well to something that we're quite keen to advocate which is that the code of practice that we have for official statistics could be adopted much more widely and be adopted by providers of information and statistics who aren't necessarily government bodies. Now there are some things in that code which are specifically designed for the government context, things like the pre-release access or the publication time at 9.30. But if you step back from those and say well why does publication at 9.30 matter, well it matters because the producer of the information, the official body, the government wants to demonstrate a commitment to its users that it will publish at a particular time come what may and regardless of political drivers. That sort of commitment to being trustworthy could be satisfied in a number of ways but the principle of trustworthiness is something that a whole range of other bodies could easily adopt and describe how they comply with and similarly with the things that we talk about around the quality of statistics, how that's explained, how you make quality transparent to users and indeed about the value of statistics, how statistics produced by any organisation answer key questions. We think there's kind of high level, we call them pillars, those high level pillars, trustworthiness, quality value and not kind of the unique preserve of government and I think it's quite an exciting debate that you've had with the wider actors in your ecosystem to think about are there ways of adopting kind of universal principles that can just help users marry up different sources in a confident way. I think that's really kind of very innovative and I'd be very happy to talk about it more actually, I think it's an exciting development. Thank you, thank you. The convener that poses this question. Well it's a follow-up from yourself I think. It's on the pre-release access to statistics, something the convener has raised I believe with the First Minister previously. You've changed your approach to ministers getting access to statistics in advance but the Scottish Government unfortunately seems to be stuck doing something entirely different. Would you advise them to change in the same way as you've done? I say that as someone who's been on the other side of the fence as a obviously a key user when I was on the monetary policy committee but at the end of the day having varied tide rules around pre-release access prevents abuse and sometimes these are accidents rather than deliberate. Somebody knows something in advance and they say something which reveals market sensitive information which shouldn't have been out there but it brings the process into disrepute and of course sometimes people may deliberately exploit pre-release access for personal gain and we think there have been examples of that in the past as well. So I think at the end of the day although ministers obviously like having early access because they can be briefed about what the numbers mean and all of this and it does mean that their initial reaction is informed rather than ill-informed so you can see the argument for it but at the end of the day I think the balance of arguments support pretty strict limitations on pre-release access. My first answer to your question is yes. A more structured answer is actually the act under which we operate as the Office for Statistics Regulation prevents us from imposing any changes to pre-release access arrangements. The act specifically says that our code of practice, this core enforcement tool cannot deal with the granting of pre-release access so it's the choice of the producer of the statistics whether they grant pre-release access. The ONS has recently chosen to remove it and actually shortly afterwards the Bank of England itself followed that. The Scottish Government has chosen another way so I can't say my yes and then enforce it. The act prevents me from doing so but I can make some kind of advocacy comments and the things that I'd advocate in this space are twofold. Firstly, why do we care so much about this? It's because at the heart of what statistics are about is that they are a public asset. They are there for the public consumption of information to understand the nature of the world, the nature of policy, the nature of the decisions being made and that vision of a public asset is underpinned by statistics being equally available to all and not partially available to some audiences and not others and by them being available as soon as they're ready, pre-release access seems to run against those two principles. Moreover, I just think that it makes the job of the Scottish Government as a producer of statistics a little bit harder. It's beneficial because it allows it to have access to ministers and to explain the content of the statistics. However, I think that you've got to set against that this perception issue whether or not, as Charles has identified, there are actual breaches. There's the perception that one set of actors, ministers, are getting a privileged access as others aren't. That means that the Scottish Government to establish its trustworthiness, used by my term again to establish its trustworthiness, needs to work much harder to demonstrate the integrity of its production process. I think that it makes their work harder. In saying that, I'm not making any judgments or accusations about the people and what they do. Genuinely, they are highly professional statisticians who do an excellent job. I just think that this makes their work harder. Just to follow up from that, do you think that the Scottish Government should be both a producer and user of statistics? Is it the case that the set-up is fine but it's just a pre-release issue that you might have a difficulty with? I'll give my first thoughts on that. I think that good statistics can emerge from a whole range of institutional arrangements. If you look across the UK, you have in Northern Ireland this model of a separate agency, Nisra, that is an agency of the Department of Finance, which operates a hub and spoke model. There's a central Nisra with about 200 people, but a lot of its statisticians are embedded into the various departments in the Northern Ireland Administration. That's a very powerful model for ensuring that those individual departments, whilst they are producing their own statistics, have a powerful professional voice. In England, you have ministerially-led departments who are very effective producers of statistics in a very high-quality way. If I could just single that one example, the work that DWP, Department for Work and Pensions, does to produce households below average income, that's an absolutely superb statistical output, which emerges from a ministerially-led department. In Scotland, you have what you might call a three-peaks model. You have a strong statistical hub in the Scottish Government, you have National Records Scotland and you have ISD Scotland who produce the NHS statistics. I don't think that I have any presumption that one of those models is superior to the other. I think that what really matters is very, very clear adherence to the code of practice, which enables statisticians to operate independently across all of those institutional settings and clear development of professional standards, a focus on quality. I think that I'm agnostic on this question of where the production takes place, if I can put it like that. However, as I go back to my earlier point, I think that it creates a little bit more of an uphill struggle in perception terms, this pre-release access point. It certainly does go beyond pre-release access, because certainly if you have production taking place within a department that's using it, potentially tensions will rise. You have to have very clear protocols, and it would be clear that the producers of the statistics in the department are independent and producing those statistics without interference. The whole structure of statistics, the ONS-Uxer and so forth, going back to 2007, has its roots in the perception. I think that it was crime statistics, if I remember right. However, some statistics and unemployment as well were basically being massaged or fiddled by the ministers in charge of the departments. That may not have been true, but nevertheless there was a perception of that. The key objective of the 2007 act was really to set up clarity of independence and production. That really needs very clear rules of operation inside the department or organisation involved. One of the things that obviously I investigated in the course of my review was whether that process was working. It was largely, but it was clear that there were occasions that some producers were being put under inappropriate pressure by their minister or the senior civil servants. You have to have a sort of robust framework there. It's possible to do it, but it needs to be clear. What happens to a few years ago? Will Selenton use it like Sir Nicholas Henderson did during the referendum? Nothing happened to him. I mean, if there's an abuse of the presses, it's up to... Wasn't he done under the code? I'm not sure that that's within your remit to decide who should be done under the code or what. It is under Mr Humpherson's. Well, we'll let Mr Humpherson answer if he chooses to on that one. I'm not going to comment on that particular case because it was some time ago. The thing I would say is that we repeatedly, both in Scotland and in other parts of the UK, will make public statements about the way statistics not only are produced by the statisticians but are disseminated and used by government departments. We have a whole long stream of correspondence where we've made some quite firm public statements, which are really reminding all who are involved with the production of government information, not simply the statisticians, that they have responsibilities to the public, and we will continue to do that. Thank you. A question from Jamie Halcro Johnston. Yes, thank you very much, convener. As islands and islands and MSP, I represent an area that is huge and diverse, and even within it there's differences within the economic speed of the local economies. I wanted to ask whether administrative data could play a role or an increasing role in providing more useful localised data within the regions of Scotland and even more local than that. Also, could that data be usable for getting an idea of regional productivity? In short, the answer is yes. We certainly have more administrative data, so one of our areas that we want to explore next is PAYE data, so we will then be able to understand in a much more granular way what people's earnings from employment are in a particular area. We could go down quite small on that one. We'd have to be very careful. We don't—it doesn't become disclosive, but we would certainly want to do that. As I said, VAT data does allow you to go down to final levels. There are some challenges around that. In your previous role, I think it was MP in Westminster who asked how much VAT was paid by a particular postcode, which I believed was a Scottish island. Actually, there was only one company trading there, and they'd had a VAT repayment, so it looked like this island was not paying VAT, but as I said, that's the function of how it's collected, and then you need to understand, given what data we have, how much can we really infer as to what's going on locally. It's not a magic bullet, but it does allow you to have a lot more granular data. We would never be able to have surveys with that level of detail. VAT data, we have 30 or 40 times as much data as we collect through our business surveys. It is really astounding the level of detail we have, but as I said, there are difficulties in how you interpret that data, but that would give you a much greater understanding of local performance. As I said, when we publish our balanced gross value added, we're looking at not just doing Scotland, but the next layer down in terms of detail. Obviously, for the Highlands and Islands, the numbers of people involved are quite small to give you much granularity or to give you much resolution, but, in theory, we can extend that further down using administrative data sources. One of the central policies of the Scottish Government, the 4i economic policy, is inclusive growth. One of the questions that we've asked in other sessions is whether there is a recognised economic definition or statistical measurement of inclusive growth, or does it fall into the category of what you mentioned before about wellbeing, an amalgamation of different measurements, which may mean different things to different people? I think that it does fall into that category of having a scorecard. One of the criticisms of GDP is that it has an average measure and it's the distribution that is important in people's wellbeing. We do try and provide more detail on the distribution of income and wealth. Those are quite difficult things to measure because you need to really get a good understanding of households in a high degree of detail. To my mind, yes, there are some definitions, some shortcuts we use, but you have to be very careful. Sometimes those shortcuts, sometimes we use in households below average income and there are poverty or low income, is measured by certain percents of the household median, but some people will say they're arbitrary, so you have to be very careful that you don't just choose one number that you have to look across the whole suite of them. Similarly for measures of inequality, the main measure of inequality is something called a genie coefficient, but that's only measuring one aspect of inequality. There are lots of other different aspects about inequality that might be salient. Also, it's inequality in what? We often focus on inequality in income, but inequality in wealth might be more relevant, and then it's also how social mobility. There are just many aspects to understanding these issues, which I think, both from a technical point of view, have not chosen one particular measure that could sometimes have perverse results through to recognising that inclusiveness, understanding social welfare, is a multifaceted thing in itself. I think that the real key here is to define your question carefully at the beginning, so don't start with the statistic and then say, well now what question can we throw at it. Think about what the question is that you want to answer, and try and define that precisely, and then go and look for the statistic that most closely corresponds to that, or set of statistics, because I think that Jonathan is right. This is territory where probably you're looking at a portfolio of indicators. Thank you. Thank you very much to our witnesses. Thank you for coming in today. I'll now suspend the public session and move into private session. Thank you.