Environmental Audit Committee - Measuring well-being and sustainable development: Sustainable Development - Minutes of EvidenceHC 667

Back to Report

House of commons

oral EVIDENCE

TAKEN BEFORE THE

Environmental Audit Committee

Measuring well-being and sustainable development

Tuesday 13 November 2012

Dr David Halpern, Glenn Everett, Nigel Atkinson, and Stephen Hall

Evidence heard in Public Questions 1 - 70

USE OF THE TRANSCRIPT

Oral Evidence

Taken before the Environmental Audit Committee

on Tuesday 13 November 2012

Members present:

Joan Walley (Chair)

Peter Aldous

Neil Carmichael

Zac Goldsmith

Mark Lazarowicz

Caroline Lucas

Caroline Nokes

Dr Alan Whitehead

________________

Examination of Witnesses

Witnesses: Dr David Halpern, Director, Behavioural Insights Team, Cabinet Office/No. 10, Glenn Everett, Programme Director for Measuring National Well-being Programme, ONS, Nigel Atkinson, Deputy Director, Head of Sustainable Development Unit, Defra, and Stephen Hall, Statistician, Defra, gave evidence.

Q1 Chair: A warm welcome to all four of our witnesses this afternoon. We have quite a lot of ground to cover and a short amount of time in which to do that. This issue is of great concern to this Committee, and it is one that we hope will attract wider public interest as the work commences, in terms of looking more closely at what the sustainable development indicators are and how they relate to policy. With that in mind, could I ask you at the very outset-we have two separate parallel consultations: we have the sustainable development indicators and we have everything to do with the well-being measures as well, and the work of the ONS-how closely these things should be brought together? How have we ended up with two separate pieces of work and how well do the two aspects of that fit together? In asking you this question, could you perhaps cast light on why this informal consultation has been going on with the SDIs, and tell us how we get wider public interest in them if they are not considered to be an absolute top priority on the Government’s agenda for this matter? I do not know who wants to put their hand up first as the start of the vibe.

Nigel Atkinson: I am happy to do so. Do we need to introduce ourselves? Perhaps we should introduce ourselves and then I can explain why we have come before you in this particular combination, because it plays to your first question.

Chair: Please do.

Nigel Atkinson: I am Nigel Atkinson. I head up the Sustainable Development team in Defra.

Stephen Hall: I am Stephen Hall. Until a month ago I was the statistician responsible for the sustainable development indicators in Defra.

Chair: Until a month ago?

Stephen Hall: Yes. I have moved in post.

Glenn Everett: I am Glenn Everett. I am the programme director from the Office for National Statistics for taking forward the work on Measuring the National Well-being programme.

Dr Halpern: David Halpern from No. 10 and Cabinet Office. I look after the PM’s interests in relation to well-being, in particular.

Q2 Chair: The introductions have shown us nicely the synergy between the two different strands of work that are underway at the moment.

Nigel Atkinson: I think so, yes. You indicated to us that in this inquiry you would like to focus first on the sustainable development indicators, mindful of the fact that we have just finished consultation on those. We were also led to believe you would want to explore the relationship between this and well-being in conceptual, architectural terms, hence the representation from different Departments today: Defra lead on the sustainable development indicators; ONS on the measurement of well-being; and the Cabinet Office and Treasury on well-being policy.

As you have indicated, we have two indicator sets. Both of those initiatives derive from the recognition that we need to have a wider set of measures to show how the nation is progressing, than simply the traditional economic sets, such as growth in GDP. We see the two initiatives as complementary. The sustainable development indicators-I will call them SDIs, if I may, because it shortens it-have a long history, as you know. They have been developed over a period of years. As I say, the last set has just been subject to consultation. They are intended to provide a high-level summary of progress across the three pillars of sustainable development: the economic, the environmental and the social, with a particular focus on factors affecting long-term and intergenerational progress. They have been selected to align with the Government’s vision for sustainable development and key Government initiatives and departmental business plans.

The well-being measures-and I am sure David and Glenn will talk about those-have emerged from a national debate and a series of consultations. They are focused primarily on current well-being and are significantly weighted towards social measures of well-being, so things such as relationship, health, community and so on. However, I think there is recognition in the well-being set that the economy and the environment have the potential to impact on current well-being. For that reason, there are a small number of environmental indicators in the well-being set, and some economic ones as well. Also, I think there is recognition that concern about the future can also impact on current well-being, so there are some forward looking measures as well. Inevitably there is a degree of overlap, but we work to ensure that the two measurement frameworks nevertheless fit together.

Q3 Chair: Do you see them coming together and having just one stream of work at some stage?

Nigel Atkinson: That is an interesting question, and one we have been discussing ourselves. At the moment, in the short term at least, we do not plan to do so. As I say, we see them as complementary but we see them as having been produced for different purposes, so the national well-being framework was not developed for the purpose of monitoring sustainable development, and the sustainable development framework does not capture the level of detail that reflects what citizens have been saying about what affects their well-being. It would be practically difficult-I doubt it could be done-to bring the two together and continue to serve the two purposes for which they have been defined. For example, if the SD framework were to incorporate the well-being measures, there is a risk it would skew that framework too far to the social, perhaps too far to the present day relative to the future. It would also make one of our key objectives in this latest version more difficult, which is to try and streamline the indicator set. On the other hand, if we were to reduce the number of well-being measures to fit into the SD framework then it dilutes the messages that have come from the well-being consultation, so there are some practical difficulties about that.

In the longer term, we are all aware that there are developments in train. There is work on valuing natural capital, there is the work that has been mandated post Rio by the UN Statistical Commission, and Glenn and David will explain how the well-being debate will evolve further. I think this is something we will need to keep under review as we go forward. In the meantime, we thought it was important that users out there had some indicator sets that they could use.

Q4 Chair: I will come on to Mr Everett in a minute but, given that there are two different streams of work going on, how do you answer the critics among the NGOs, for example, that that is giving out contrary signals in respect of policies that need to be adopted, not just within Defra but right the way across Government Departments?

Nigel Atkinson: All I can say is that we think they have different but consistent purposes, so the fact that they are saying different things does not make them inconsistent. The sustainable development indicators are high level. They are taking a panoramic view of sustainable development and the development of our nation as a whole. As for the well-being measures, there are more of them. They are much more specific. They are focused particularly on the social, including a greater emphasis, given the SDI set, on subjective measurement, and they can perhaps be used in different ways and certainly for different audiences, so as I say they have different purposes.

Q5 Chair: Thank you. Mr Everett, do you want to respond?

Glenn Everett: Certainly. It might be worth giving a bit of background about measuring national well-being, which I lead. It is almost two years since we started the programme and, during that time, we have consulted quite heavily with a range of users from the general public right through to corporations and different Government Departments. We received something like 36,000 responses to our development work. What we did then was to distil the common messages into what we now propose and now publish-albeit, on an experimental basis-as 10 domains with around 40 indicators that reflect, as headline indicators, national well-being.

The framework that we use stems partly in a similar way. The top level framework uses the three pillars of environment, economy and social, but we have distilled that into 10 quite clear domains and at the centre is an individual’s well-being. Outside of that there are the actual factors directly affecting the person’s well-being, things such as health, where we live, what we do, education skills, relationships, and their own finance, and in that there are the contextual other domains. The three other domains are the economy, environment and governance. As dimensions of this, we recognise the sustainability issues, as well as equity, to try and describe all of this. However, I stress that it is early days in some of this development.

It also has an international element. The UN are very interested, and the OECD have recently issued some guidance on measuring subjective well-being to try and aim at international comparability. The Europeans are developing quality of life measures. Let us say we are quite influential, in that they are trying to adopt some of our developments and our measures in taking this forward. It is complex and it is multi-dimensional, and we are working quite closely with Defra to try and make sure that there is limited overlap-and where we do overlap it is understood-and to take this work forward.

Q6 Chair: Do you see there being eventually perhaps one single stream of work, or not really?

Glenn Everett: I am not sure. It is something we need to keep in mind, but it may be sensible to keep them apart at times. To me it is too early to say one way or the other. I would say reviewing it is something we can do in a couple of years’ time, to see how it does properly overlap. From our development side on the programme, we want to look again at the framework, look at the domains and measures to make sure we have it right. We continually consult on some of this. We use technical advisory groups. We work with the different policy Departments to understand their requirements as well, to make sure we are getting it right so to speak, but I see in the future we will look at the framework to see if there is a sensible way to do it. In the meantime, our aim is to always make it easy for the user to go from one to the other, so if they are drilling down into our measures, for example, sustainable development, they will be signposted directly across to the Defra side.

Q7 Chair: Finally, how much interest is there from the Treasury in terms of the work that you are doing to develop this stream of work?

Glenn Everett: From our point of view on the Treasury, two years ago the Prime Minister asked the National Statistician to take this work forward. We originally started it as about a five-year development programme. We were ring-fencing some of the programme costs to take this work forward. My expectation is around 2015 we will want to be mainstreaming this information, but in the meantime we will be doing a lot of the developmental work and measurements on things like human capital, social capital and natural capital. We are not doing significant amounts of new surveys. There are four new subjective well-being questions we now ask, but they have just been added to an existing survey. So the costs are quite small in the big scheme of things, and I think Treasury have recognised that sort of work and how we are taking it forward.

Q8 Zac Goldsmith: This is a question for Mr Atkinson. Could you tell us, broadly speaking, how you envisage the new SDIs being used in practice, and perhaps how that would compare to the usefulness of the SDIs they replace?

Nigel Atkinson: We know that the previous set had a very wide use-so, beyond Government-particularly in education, a third sector in business, and indeed in local and regional government. We know there was a wide demand for them in that sense, and we want to make sure that the revised set that we produce remains fit for those purposes. That is important to us, and it is one of the reasons why we will be looking very closely at the feedback we get from the consultation.

As for the set that will be used in future within Government-and I think we have to bear in mind this is a headline set, this set of indicators is quite high level-we see the value of them being potentially to encourage the right sort of discussion, at the stage at which polices are being developed, about what sort of impacts one might want to be mindful of when designing policy, or at least where potential gaps are-

Q9 Zac Goldsmith: On that point, are there signs that that is already happening, practically speaking? When decisions are made by different Departments, which would have a bearing on the new SDIs, do you see these SDIs having an influence on those decisions in Departments other than Defra?

Nigel Atkinson: We will have to see whether that is the case and, of course, it is always difficult to measure these things. I think that is the intention, that it would encourage Departments to have regard to the breadth of the possible impacts of polices they are looking at and where the gaps are, potentially. What becomes more difficult is using the SDIs after the event to evaluate the impact of a particular policy. One of the difficulties with that is that most of these indicators will be potentially subject to a wide range of factors that will influence them, so determining a hard, quantitative, causal relationship, between an individual policy or even a potential group of policies on a particular indicator, is going to be quite a challenge.

Q10 Zac Goldsmith: Before I move on, can you give me an example of how practically the SDIs will be of use, as you said the previous set were in relation to business? You said there was great enthusiasm from business for the first set.

Nigel Atkinson: We know that businesses are developing their own sustainability strategies, and some businesses are very much in the vanguard of that. I think some of them have looked at the indicator set we have had, as a starting point or a sense check against their own direction of travel. That is my feeling. I would be hard pushed to prove that.

Q11 Zac Goldsmith: If Mr Hall wants to jump in then please do, but I want to ask: the ONS is leading on well-being. What would be the pros and cons of your organisation or your Department also managing SDIs?

Glenn Everett: It is not a simple question.

Zac Goldsmith: No, it is not. We touched on this earlier, I suppose, in relation to-

Glenn Everett: Yes. It is early days. At the moment we are developing measures of well-being, as in trying to understand more complete measures of how the nation operates. The background is often the criticism that GDP masks a lot of issues beneath it, so we are looking at, not throwing away GDP but looking at supplementing it with other measures to better understand what is happening going forward. The sustainability issue is there as a dimension. It is one where we need to develop further, from my point of view. How it actually works long term will depend on quite a lot of things, including the structure of government and how we work as a Government statistical servers where people across different Departments will work under the same code of practice.

Q12 Zac Goldsmith: Is there an argument that, if it with was the ONS, the public would have more confidence in the objectivity of the criteria and also the data that are generated?

Glenn Everett: It could be a perception. The reality would be, if it was done under a Government statistical service code of practice and became a national statistic, it should have the same weight. The perception should be that if ONS were doing it, we would be-I hope-perceived as being more independent, but the reality is it should be of the same quality.

Q13 Zac Goldsmith: Do you want to add to that?

Stephen Hall: Yes. The previous set was designated as national statistics, and was produced within the code of practice as if it was within the Office of National Statistics. We have a dispersed statistical system in this country, so statisticians in Defra are as independent as statisticians in ONS. We have never had anyone questioning the way in which we were undertaking the production of the statistics behind them.

Q14 Zac Goldsmith: That applies to the new SDIs as well?

Stephen Hall: That is the intention, yes, but they will have to go through a formal assessment by the UK Statistics Authority in due course to continue to have that designation.

Nigel Atkinson: It is worth adding perhaps that, in almost all cases, the indicators we are using are based on existing data sets, so these are data that have already been produced. They are not data that have been produced specifically for this purpose.

Glenn Everett: Some of which are produced by the ONS.

Q15 Zac Goldsmith: The final question I have is to what extent is the choice of the SDIs driven by the need to also submit data to Europe for their version of SDIs?

Chair: Mr Hall?

Stephen Hall: There is a set of sustainable development indicators that is maintained by Eurostat, but that is designed to make best use of data that Eurostat has access to. There is no reporting requirement for us to report sustainable development indicators to Europe. However, because we have to provide the data for other purposes to Eurostat, a lot of the data that they are drawing on we have, and we will have it as part of our own indicator set, so there is a lot of commonality between what is happening in Europe and what is happening to our own indicator set.

Q16 Neil Carmichael: Good afternoon. Targets and SDIs. There do not seem to be a huge number of targets. Is that a general feel that they are not necessary?

Nigel Atkinson: I should think, as far as the SDIs are concerned, the SDI framework is not a target setting framework, and the SDIs do not themselves introduce new targets. In our approach to this, we have reflected the Government’s approach to transparency and accountability, so the guiding principle area is not, for the most part, accountability through a centrally designed system of targets and processes but rather to focus on being open and transparent with the data, so that the public is informed and able to scrutinise performance.

Having said that, some of the indicators that have been selected are already measures in Departments’ business plans, and one or two of those already have targets with a small or large "T" attached to them. In a sense, those are the issues that Government has decided it wishes to set targets for. The SDIs themselves, and that process, is not setting new targets but no doubt when a reader looks at the SDI set no doubt they will want to have regard to the targets that are there already.

Q17 Chair: May I just interrupt? Can you name two of the Department business plans where this is actually happening, if I may cut across Mr Carmichael there?

Neil Carmichael: That’s all right.

Nigel Atkinson: Yes, I think there are two direct read-across, so the indicator on greenhouse gas emissions links directly to the DECC business plan and to the targets under Kyoto, so that is one. The other one is in Defra’s business plan, so you will see we have a headline measure of wildlife and biodiversity, which is bird population. One of those indicators for farmland birds is in the Defra business plan with an objective of reversing the downward trend. Those are examples where those targets exist already in Departments’ business plans.

Chair: Would that extend to Ash? Sorry, Neil.

Q18 Neil Carmichael: We discussed that in detail yesterday. Transparency, certainly, absolutely right, and Nigel mentioned that in his answer to my original question. How are householders likely to be able to judge their own performance if they do not see others using targets?

Nigel Atkinson: Sorry, I am not sure I follow that. I am not sure entirely how householders will be able to-

Q19 Neil Carmichael: You are talking about the public?

Nigel Atkinson: Yes. I am not sure that householders will ever quite be able to use the SDIs to determine their own-

Q20 Neil Carmichael: I am not suggesting to use the SDIs, but if you are not using targets, in the absence of any targets or at least enough targets, depending on your view, how can other people get a feel of where they should be going?

Nigel Atkinson: As far as the indicators are concerned, what we will consider doing is adopting a system of traffic lights, so that it is at least possible to see what the direction of travel of some of these things is. That would be a fairly simple process. It will not be a system of targets. If the feedback from the consultation is that respondents would find that helpful, then I think that is what we will consider doing.

On the question of targets, ultimately it is a matter for Ministers what they do and to decide if they want targets.

Q21 Neil Carmichael: What about research and development, because you mentioned that as an important area? Where do you think research and development sustainability needs to get to?

Nigel Atkinson: As I say, there is no target for the measure so if I was to give you an answer to that it would imply a target.

Q22 Neil Carmichael: Does anybody else want to comment on that?

Dr Halpern: On targets and indicators or do you mean more specifically-

Neil Carmichael: I am talking about research and development. I am moving on to that area in terms of sustainability.

Dr Halpern: I think it does relate to your earlier question and, as far as these headline indicators, remember, you can have an indicator or a measure without it being a target. Nonetheless, it is going to be pretty important, in terms of the signals they create. There are certainly some big questions that lie behind this whole conversation, which are: how do these various things interrelate and what would be the implication, and particularly, what could we do differently that might impact on these things? Those are really important questions for us to have an eye on. Having some reasonably high profile indicators out there do help to frame the research agenda going forward. That is one reason why it is important that it be flushed out in this kind of way.

Q23 Neil Carmichael: By that answer we are talking about indicators rather than targets?

Dr Halpern: If a target means that you have something where there is a sanction that follows if you do not meet it, yes. You can have your odometer in your car and, even if you do not have a target for what it is at, it is pretty useful. It is giving you important information.

If I could go back a bit for a moment to some of the earlier, bigger questions, I think one of the things that frame this is that there is a near universal understanding that there are serious inadequacies with some of the headline measures, including GDP. We do not have a target for GDP, but we know it has some gross inadequacies of things that it does not cover and it is a seriously deficient measure, right? What else would you add to it? Clearly, Defra has a long history of being an environmental conscience as well as an activist in policy terms around the environmental agenda. It is just that we know the inadequacies of GDP are far more wide-ranging than that-as flagged up by Stiglitz and others-and the well-being agenda is one of the things that has been brought in to that. All kinds of things are not captured by that, such as kindness, compassion and caring that bear on our well-being enormously. There are lots of big policy choices that are made where, self-evidently, we may not be maximising GDP. By deciding we are going to concrete over the entire rest of the UK, it might be pretty efficient in GDP terms but there are lots of other reasons, both environmental and well-being, that you might not do it. Not having an indicator is a strange situation, where we have a set of gut instincts. Often our senior Ministers, indeed, Prime Ministers, are making judgments and yet we have an indicator set, which does not correspond to what seems to be how they are making judgments. I do think it is a fundamental issue that has both policy as well as research implications.

Q24 Neil Carmichael: GDP growth has to be sustainable, and concreting over the whole country would probably leave us quite hungry, so it has to be all balanced together. We have binding targets in some areas, though, emissions and air quality and so forth. Do you think they are appropriate? Who would like to answer that question? Glenn?

Glenn Everett: Appropriate for what, I am sorry?

Neil Carmichael: This is a discussion about whether or not we should have targets, basically. We have them in air quality, for example.

Glenn Everett: I think it is most important that we have the information. My primary concern is that we have the evidence on which to base anything on; that we have a solid piece of evidence that has developed over time, consistently with consistent methods. To say there is a target on air quality is quite important. I am not in a position to judge whether it is appropriate or not-just as a GDP target may not be appropriate-but what is appropriate is that we have good measures of those things, so that if others put a target on it we can be confident that it has not been fiddled, for example.

Q25 Neil Carmichael: At Rio+20 earlier this summer, it was agreed to develop a set of sustainable development goals, which I think is pretty reasonable. To what extent might your SDIs have to be revised to accommodate any new sustainable development goals?

Nigel Atkinson: They may well do. As you say, Rio agreed to develop sustainable development goals and there is an open working group that is being established to take that work forward.

Q26 Chair: Do you say it is established or is being established?

Nigel Atkinson: I have "is being established", but I-

Chair: It is not established yet.

Neil Carmichael: I think it is a job in completion, is that right?

Chair: Work-in-progress.

Neil Carmichael: Work-in-progress that is it.

Nigel Atkinson: Let me come back to yon on that one. Alongside that, the outcome called on the UN Statistical Commission to launch a programme of work to develop broader measures of progress to complement GDP, in order to better inform policy decisions. I understand that has been asked to take account of existing work, and when the Commission takes that work forward we will be looking to engage with them to progress it and to ensure best fit with the indicators we have or, indeed, to assess whether the indicators we have need to change in the light of the direction of travel of that work, so that it is something we will be engaging on.

Q27 Neil Carmichael: Are you comparing the trusting indicators with other nations, in terms of the way in which they are operating?

Nigel Atkinson: International comparisons are always difficult. Stephen, do you want to talk on that one?

Stephen Hall: Yes. We have tried it in the past but find it very difficult because of inconsistencies in the data across countries, to the extent where you have so many caveats you cannot compare one country to another. At the European level Eurostat do that anyway, so where they can produce consistent indicators, they are available, and available to the public to evaluate. We found it was too difficult for ourselves to match our particular indicators to international comparisons, so we will be relying on the work that Eurostat has done.

Q28 Neil Carmichael: How often would you engage in a dialogue about those issues?

Stephen Hall: You mean in policy terms?

Neil Carmichael: Yes.

Stephen Hall: I defer to Nigel’s opinion.

Q29 Neil Carmichael: Stephen’s answer is you are looking at the issue. It is complicated, and I recognise that because of different ways of measuring all sorts of things, but is there a constant dialogue or an occasional dialogue with various nations that we ought to be talking to?

Nigel Atkinson: I am not sure I can answer that question. I am not sure there is a forum that regularly takes this forward, but it comes up in a number of places infrequently I would say. No doubt the work of the UN Statistical Commission will be a focus for bringing together the question of how you measure these things, including internationally.

Q30 Neil Carmichael: Would you describe the UK as in a leadership role or in a reactive role in this?

Nigel Atkinson: I hesitate to say this, but I am told we are rather in the vanguard or have been. Glenn certainly seems to think so in respect of well-being, and I think it is probably true of SDIs.

Dr Halpern: For well-being, if not for the more specific environmental measures, I would say, self-evidently, we are in a leadership role in the UK. There are lots of meetings that are occurring on OECD. I know the National Statistician will be going to meetings all the time to consider issues, as well as the specific question of what is a reasonable measure of subjective well-being, which is a more focused measure and was also identified by Stiglitz as being a gap, so a very basic question. If we use a 10 point scale to ask people how satisfied they are with their lives, and New Zealand use a five point scale, or someone else uses something called the ladder of life, there are lots of different simple measurement items and clearly it would be far better for countries to end up with a similar set of measures.

The National Statistician has also been pursuing this broader issue of a balanced dashboard of measures of the state of the nation and economy and it will be a long haul to get other countries to a similar set of measures. Clearly, it would be superior if we could get some convergence across countries where there is comparability in the measure. There would be much more value for us. I cannot speak for all areas. I can definitely say, on the subjective well-being side, we are ahead of most other countries and we are taking a leadership role.

Neil Carmichael: Thank you.

Q31 Chair: Do you want to come in, Mr Hall?

Stephen Hall: Yes. On the sustainable development side, I would like to similarly claim that we have been in the lead for many years. We were very much involved in the formulation of the EU sustainable development indicator set, and I personally was involved in working groups that developed the EU set of indicators. We have also been very extensively involved in a number of international initiatives on sustainable development indicators and are often invited, because of our expertise and our leadership, to help influence international work. In particular, the United Nations Economic Commission for Europe, OECD, Eurostat are doing a joint project to look at developing a new conceptual framework for sustainable development indicators, and we have been involved in that and, in due course, that would hope to move towards greater harmonisation and comparability across countries.

Q32 Chair: They are current as well?

Stephen Hall: Yes.

Glenn Everett: I was going to add a little bit to what David has mentioned. He mentioned our influence and everything else. I mentioned earlier OECD have recently provided guidance to all their members on measuring subjective well-being. One of their recommendations of what they should use is a direct lift of the UK question on life satisfaction, with the 10 point scale. It was a feather in the cap of the UK that they had adopted it directly, and reflects our influence. There is a range of things happening worldwide, ranging from the work of the UN, OECD, Eurostat, and myself, and the National Statistician recently went to an OECD world forum. We have regular meetings-I was in Luxembourg last week-about quality of life indicators, which reflect very much what we are doing. We are invited as experts to these, these are not "for all" meetings. I want to highlight that there are a range of things in development and we approach it multi-level ways. We use different people to approach, as best we can, whether it is sustainable development, whether it is well-being, whether it is quality of life. There are all these indicators we need to influence.

Q33 Chair: Just before I move to Dr Whitehead, can I go back to something you said just now, Mr Atkinson, about the targets? We were looking at whether or not targets might be useful in all of this and, given that the consultation has now closed, is there any feedback from the consultation on the subject of targets and, I suppose, linking targets to achieving legislation and so on, that you would wish to share with us?

Nigel Atkinson: Yes. At the moment we are still analysing the responses to the consultation, so I cannot give you a comprehensive assessment of what it is telling us but, generally speaking-and I am looking here at the results from the online survey we ran, where we can have a quick look at how the questions were answered-I would say there does seem to be a very high measure of agreement on the issues that have been selected for the headline set, and on the indicators we have chosen for measuring those. Inevitably, the indicators have a lower satisfaction than the issue itself, but are still very high. However, there is very little appetite for substituting any of the headline measures for any of the supplementary measures.

Q34 Chair: I was interested specifically on targets.

Nigel Atkinson: On targets, yes. However, there are some themes that seem to be coming up as things that people would like us to look at. One of those is the setting of targets. I do not know how strong the feeling is about that yet, but it is certainly something that we see. There are a few others as well that, doubtless, we will pick up as we go through the discussion, but, yes, there are some people who would like us to set targets at least.

Q35 Chair: A majority?

Nigel Atkinson: I do not think I am in a position to know at the moment because we have not got that far.

Q36 Chair: When you do know it would be helpful if you could share that with us.

Nigel Atkinson: For sure.

Chair: Thank you.

Q37 Dr Whitehead: In the old set of SDIs there was a specific indicator that focused on inequality. Where do you think the question of inequality is addressed in the new set of SDIs, and how directly?

Nigel Atkinson: In the current set, inequality of different types features in a number of different elements. For example, the fuel poverty indicator is related to income inequality; the fruit and vegetables indicator gives data by income band; infant health by type of occupation; smoking and alcohol differentiates between men and women, so there are those sorts of things. What we have not done-and this is probably what you are alluding to-is focus to a greater extent specifically on environmental inequality, so yes, in the last indicator set there was a first attempt-an experimental attempt, I guess-to define a measure of environmental inequality that was able to tell us something about the relationship between deprivation and environmental conditions.

That was a composite indicator. It was attempting to pull together and weigh a lot of different environmental impacts. Because of that it was certainly data hungry and quite resource intensive to produce, and some of the data that fed into that composite measure are no longer available. Some of those data series have been discontinued, for example, data on green space. I guess we could expect that sort of problem to re-emerge if we were to try to produce this as a regular time series. We have had to balance the resource burden and the data challenges associated with continuing to produce it, with the added value we think we are getting from it. In the light of that we decided not to include it. I said we seemed to be getting some themes from the consultation responses, and indeed one of them does seem to be the suggestion that we have not paid enough attention to inequality, specifically environmental inequality. Subject to having a proper look at all those responses, that is something we will want to go away and think about again, and certainly, if the Committee have views on that, we are more than happy to take those into account as well.

Q38 Dr Whitehead: You mentioned that some data sets have disappeared. On the other hand, there are a number of data sets that could perhaps be teased out to a greater extent, as you say, in terms of environmental inequality, such as things like healthy life expectancy, designated as a variation across the country, higher education, better salaries, that sort of thing. Is that something you are intending to look at? I think it is something you mentioned or-

Nigel Atkinson: Yes. As I say, we will need to look at this whole area again, in view of the responses we are getting. In principle, of course, many of the measures we have are averages, not just in relation to the environment but in relation to all of these things, and it always begs the question, "Well, what is the distribution around the average?" That is a very good question, and there is very useful information you would want to have there in almost all cases. The question is how do you manage that level of data and still produce a set that is streamlined? It may be that the way forward is not to try combining this all in one indicator, but somehow to signpost from the headline finding through some analysis of distributional impacts. I do not know, but that is something I think we will need to look at.

Q39 Caroline Lucas: On the back of that-and to test the water to see how strongly the issue is felt around inequality-I would like to say that, notwithstanding I appreciate the extra difficulty in managing the amount of data, there is so much evidence out there now, levels of income, the distribution or whatever it is. It is the relative side of that, not just the absolute side, which is particularly corrosive. Looking back to things like the London riots or whatever, that difference between the best and the worst actually matters. I think trying to use an average risks masking something if inequalities are rising. We know all of the work from Kate Pickett and Richard Wilkinson, and everybody else, about relative inequality being such a corrosive issue, so a way of capturing it, although I accept it is more difficult, is very valuable.

Nigel Atkinson: We are certainly open to ideas as well, so what you just said on that will be useful to us. We also have to think about prioritising as well, so any suggestions on how we manage the volume of all of this would be helpful also.

Chair: Moving on, Peter Aldous.

Q40 Peter Aldous: First of all, could we look at the measures that have been dropped? How were decisions made on which of the existing SDIs to drop from the new set? I would welcome an explanation on that.

Nigel Atkinson: As I said, the intention was to streamline the indicator set this time. We wanted to do that because we were very mindful of the difficulty of trying to interpret, in some sense, overall outcomes from a very large number of data sets. I would say there is indeed an international trend in that direction as well, more generally. We started with the aim of streamlining. That involved prioritising. It involved necessarily dropping some things, as you have identified. Where we have dropped indicators, i.e. we have not chosen an indicator from the old set, generally speaking, it is either because I think we have identified a higher level indicator that will serve the same purpose, or effectively and sensibly cover the issue at a higher level, or we have dropped it because, in principle, the issue is covered in the national well-being set, so essentially we have carved out the space for ONS to cover those issues in its data set.

Q41 Peter Aldous: Were Ministers at all involved in the process of which indicators to drop?

Nigel Atkinson: There is of course the usual machinery of Government process for agreeing indicator sets for consultation, but I would say that on this sort of issue there is a very wide measure of agreement across the piece-including out among stakeholders-about which issues are important to sustainable development. It is fairly straightforward. I think the discussions we have had have largely been about how you best measure these things, given inevitably imperfect data. It has largely been a technical discussion, I would say, led by statisticians for the most part.

Q42 Peter Aldous: You mentioned consensus across stakeholders. When the consultation exercise was carried out did you get any feedback from the public, from people saying, "Hang on, you have got it wrong in dropping this particular indicator"?

Nigel Atkinson: Not so much dropping indicators, in the sense that, as I say, there is a very high measure of agreement about the bits that we have looked at-I have to keep qualifying that because we have not looked at everything yet-in terms of the issues being the right ones and, on the whole, being measured in the right way. To the extent that people have suggested to us that there are other things we should look at, or perhaps have not paid enough attention to, targets and inequality are the ones mentioned. Other things that have been mentioned are transport, although we need to get to the bottom of why and whether that is really about environmental impacts of transport or about social issues associated with transport. Neither of them is clear at the moment. Youth unemployment has been suggested as something we should perhaps have placed more prominently. Indeed, the discussion that we had at the beginning of this meeting about whether there should be a combined framework for well-being sustainable development is also something that crops up. Those are the sorts of things so far that we have identified as being fed back to us.

Q43 Peter Aldous: I can imagine that youth unemployment is a sensitive and a high profile issue at the moment. Did it surprise you that transport was brought up or not?

Nigel Atkinson: It was probably brought up because it featured to a large extent in the previous set. Would that be fair?

Stephen Hall: Yes. There were more transport-related indicators-

Nigel Atkinson: Probably the submission has been noticed, so the question is being asked about why. I think that is where that is coming from, probably.

Q44 Peter Aldous: If we look at particular measures, such as the economy, there are obviously four headline economy indicators of GDP: unemployment, poverty, knowledge and skills. Then there are various supplementary indicators. I would be interested in the role that the Treasury played. Did they influence the selection of SDIs on the economy at all? Did they play a key role?

Nigel Atkinson: The economic measures, the measures that Treasury lead on, were proposed by Defra to Treasury. Treasury accepted them.

Peter Aldous: Defra suggested them and Treasury said, "Fine, that is okay".

Nigel Atkinson: Yes.

Q45 Peter Aldous: If we look at the supplementaries, there is something called public sector debt in the indicators. Could you explain why this is one of the indicators and what it is trying to show in terms of sustainable development?

Nigel Atkinson: We consider debt to be important for sustainable development because it has the potential to impact on future generations, so it is potentially a big issue in relation to intergenerational sustainability. We have included it as a supplementary indicator, because we think it is useful to inform a reading of several of the headline and other indicators and not only the economic ones. I guess we might also want to think of it as a leading indicator, in the sense that it might be used by the reader to inform a judgment about whether some of the current values for the other indicators are sustainable in the longer term. We think it is there to help present a rounded picture of whether the headline economic measures-and indeed, some other measures-are sustainable in that sense.

Q46 Peter Aldous: Did you consider that there is any level of debt that should be reached when it begins to constitute a burden that is unsustainable for future generations, and did you have regard to whether there needs to be a threshold when warning lights begin to flash?

Nigel Atkinson: I think the Chancellor has some fairly public plans in relation to debt reduction, and we did not think it wise to suggest alternatives to that.

Q47 Peter Aldous: Looking at the issue of Greece-obviously, that is the bête noire-we see unsustainable levels of public sector debt there, and that non-sustainability has very much been represented in the form of high interest rates that the Government has to pay for its borrowing. Why not have the cost of borrowing, the bond issue interest rates, say, as an indicator?

Nigel Atkinson: Instead of the stock debt?

Peter Aldous: Yes.

Nigel Atkinson: I can see an argument for that. I do not know whether we had that discussion or not at the time.

Stephen Hall: I do not think we did.

Chair: Sorry, I could not hear that.

Stephen Hall: We did not have that discussion. This was a measure that we agreed with the Treasury and the discussion did not include interest levels.

Q48 Peter Aldous: It might be something worth revisiting?

Stephen Hall: Yes.

Peter Aldous: Thank you.

Q49 Neil Carmichael: Can I talk a little bit about the Behavioural Insights Unit because, first of all, I am not clear what it does and I would be grateful for an explanation?

Dr Halpern: Basically, that is what most of my time is spent on rather than this. Essentially, with a more sophisticated account of what drives individual human behaviour and decision-making, the idea is that, with that in hand, you can do better policy, policy that is cheaper, easier, better for citizens and so on. That has been our role in Government. It is almost a cross-cutting approach to policy, asking, "Is there a better way?" It might be less regulatory and less costly, which works better, and I would say it has been extremely successful.

Q50 Neil Carmichael: By definition, clearly, that is cross-departmental. It has to be, because you do not want silos with different kinds of behaviour or you do not want situations that are blocking your progress. Do you find working with Defra, for example, relatively easy, given your location in the Cabinet Office and No. 10?

Dr Halpern: It does require cross department working. To some extent we are very sensitive to the PM and DPM’s priorities, so it is very much focused on growth in employment and so on at the moment. We are a small operation but, in principle, we can roam around quite widely and we do so. We also support the growth of behavioural knowledge and expertise across Government as well. An example related to environment would be-we probably work a bit more with DECC than we do with Defra most of the time-where we have shown that you can encourage people to insulate their homes by approaching the offer in a different way. The classic response is, "Well, let’s offer a bigger subsidy" but the interesting question-which is a no-brainer anyway-is: "Why doesn’t everybody insulate their home since they will get their money back for their loft in a year anyway?" When you look at it more closely, sometimes what you find are quite prosaic but important details. The reason why people do not insulate their loft often is because they have stuff in it that they should have thrown away 10 years ago when they moved house last time. Instead, if you make an offer of a loft clearance scheme, even with no subsidy, you get a threefold increase in the uptake of loft insulation.

Those are the kinds of things that we do. Sometimes it is very nuts and bolts. Sometimes it goes back to core policy. It also explains why we are interested in well-being issues and why I will look after it long term, because the well-being agenda takes you into some of the same space. You have to say, "Well, why do we make the choices that we do? What is it that drives it?" Often, if you look at the literature, you find that people often misremember what made them happy, they mis-predict what makes them happy, and so on, and so you get caught in the same current of issues around how people make decisions, if that makes sense.

Q51 Neil Carmichael: That sounds good news for clutter consultants, but I suppose well-being is all about your sense of education or the feeling you have about yourself, self confidence, and then I suppose health. Well-being is job opportunities and so forth. Presumably, within your remit, you consider all of those things, and the question, I suppose, is: how do you pull those levers together to do the things that they should be doing, or might want to start doing, in terms of protecting the environment?

Dr Halpern: You can use what we know from behavioural economics without reference, well-being if you want to, and just say that we have a very specific issue, such as we want to improve the tax take, or we want to make people more respectful, recycle more, and you can focus on a very specific issue. You can do that. One of the reasons I think that well-being is of great interest-and is arguably a very special indicator-is that you can flip the question on its head and say, "Well, if our role as policy makers is to try to improve population well-being, not just for now but for the future, what are the drivers?" It raises the fundamental question, what are the drivers?

Clearly, it is self-evidently true that some things, such as if you do not live in a healthy environment, affect well-being. It flags some things that you would expect, like income; inequality also kicks in, issues of personal control, but lots of other variables, which, historically, tend to have been more neglected also come into play. Social capital is a good example, as is how people treat you in everyday life; if they are rude to you, it is extremely unpleasant and has disproportionate impacts, whereas that tends to be neglected. This insight does have very practical policy implications. If you have an issue such as keeping open post offices, where it looks like there is a massive subsidy that runs through it, policy wonks might highlight that there are cheaper ways of doing this, such as direct transfers and so on, rather than spending £400 million. But people often have an intuition-you may do yourself, and your constituents-that something happens with a post office that is not captured by the current set of statistics. It is genuinely the heart of a community, people bumping into each other, and so on. Our dilemma is that you are making judgments on a set of instincts, as I said earlier, where the indicators do not seem to be capturing some of those things. Well-being essentially flushes some of those out. The broader ONS work starts chasing down what are a fuller, more balanced set of drivers that account for-in a deeper sense-people’s well-being. It takes you to different marginal choices, it switches on the light to a whole set of variables that you did not see before.

Q52 Neil Carmichael: Within that, clearly there are a huge number of policy choices. The obvious one has to be-are we going to go for economic growth first and foremost? I was struck by a comment earlier about the concreting. I thought that was quite interesting. First of all, if you concreted over the whole country you certainly would not be protecting well-being, and you would be putting long-term economic growth at risk anyway because of a number of factors that would not work out. The policy choices that you are making are quite critical, therefore the real question is: what are your key indicators?

Dr Halpern: The reason I am saying the well-being variable stands out is because embedded in it, precisely, are those empirical questions. As you said, it is generally obvious, although sometimes it will not be so obvious, that for most people concreting over all of Britain will not increase well-being, right? But it is not obvious from a GDP point of view that that would be the case. This is about driving you to a more balanced set of indicators. The fascinating thing about it is that if you are a statistician you have a huge problem, you have a data reduction problem, there are a million different variables you could look it, and it is just beyond our capacity. Yet, in everyday life, as individuals, we are making real choices. Shall I buy a bigger house and take a longer commute? Balancing income and the size of the house, and how long I spend with my family means that, ultimately, a choice has to be made. The implication is that we are comparing apples with pears in some decisions in everyday life; well-being measures in principle gives you a similar kind of tool for making decisions in policy terms. When you see this array of variables, you start to be able to say, "Actually, what would be the marginal benefit of a bit of this or a bit of that or a bit of crime reduction on well-being?" It gives you a common metric, which is very powerful.

We are still at early days to the extent where that is truly possible, but that is, in some sense, the ambition behind what the national statistician is doing. Across the world, quite a number of policy makers would be interested: does it give us a yardstick to enable us to compare apples with pears?

Neil Carmichael: Thanks.

Q53 Caroline Lucas: Following on from what Neil was saying, and apologies if you covered this before I came in, but you were talking just now about how economic growth might be traded off against other things that you might decide are more valuable in terms of leading to well-being. Does the work that you do ever lead you to come to a conclusion, or point to the fact that beyond a certain level of economic growth the marginal benefits are not actually delivering well-being? What I am driving at is that at the moment everybody assumes that economic growth is the be-all and end-all of our economy, and indeed our whole way of life, that is what we are trying to do-maximise GDP. Yet, if we could demonstrate that in some ways that economic growth is not delivering what we think it is, because of the associated environmental costs or the fact that it means we are working a million hours a week and we are all terribly stressed, where does that become visible in what you are doing?

Dr Halpern: We can answer that question empirically. There is a separate question about the politics and the decisions you make around that, but we know, both cross-nationally and within countries, the relationship between income and life satisfaction, for example, is curvilinear. It looks like it is diminishing returns in the way you describe, although the correlation is also quite respectable in its own terms. Equally interesting is when you look along a line of different countries and ask why it is that Denmark is way above where it should be for its level of GDP in terms of life satisfaction? Or France is way below. What else is happening? Sometimes, in some countries it could be specifically environmental, but also it is other factors. An example would be social trust; do you think people can be trusted? That is a key variable in terms of bearing on your well-being. Just to make a final point more immediately relevant to you, it does flush out some important marginal trade-offs. A tree may absorb a certain amount of carbon and house a certain number of birds, or whatever, but in Defra terms they may not really care about where that tree is-it is neutral- whereas in well-being terms it becomes pretty clear that when people can see a tree, it has noticeable, positive and demonstrable effects on your well-being. It changes marginal decisions in an interesting and intriguing way, in a way that, frankly, we have only just begun to explore this point.

Q54 Caroline Lucas: This is an interesting thing that I just discovered on your social mobility indicator, you are looking at the proportion of people who have jobs in higher level occupations than their father, which seems mind-boggling in a number of ways. Why is that important in sustainable development terms?

Nigel Atkinson: I would not set too much store by the term "higher level occupation", it is a statistical term. I can see how that terminology might be taken to reflect a judgment on the social value of a particular occupation.

Chair: It does, yes.

Nigel Atkinson: Actually, what it is really trying to do is make a distinction between the earnings capacity of different types of occupation. The reason why we want to differentiate in that way is in order to identify the extent to which higher earning occupations tend to be the preserve of certain sections of society. Patterns of inequality and lack of social mobility can be imprinted from one generation to the next, so it is a key issue for inter-generational well-being. I would not set too much store by the label that that is the essence of what it is trying to look at.

Caroline Lucas: I like the explanation better than the label, I have to say.

Nigel Atkinson: We will think about changing the label if we are allowed to do that.

Q55 Caroline Lucas: The new social capital indicators are still to be identified, and where will the data come from for those indicators now that the Audit Commission surveys have been abolished?

Nigel Atkinson: We have had to say, "Watch this space" because it is a hugely different thing to crack, and no one has done it yet. We are clear why we want to do it-sustainable communities and institutions are the foundation of a strong society, which in turn underpins sustainable developments. It is something that we think is really important to try to measure but, as you will appreciate, that concept has very difficult measurements associated with it, to the extent that if anyone does measure it, it is often represented by property measures and combinations of several proxy measures. It is very difficult to identify a single headline summary measure that is sufficiently representative. It is something we need to work on; there are some possibilities with the Cabinet Office Community Life Survey that we may be able to look at. Of course, we are very conscious that ONS is looking at this from a well-being point of view as well. We will want to make sure we are joined up with thinking there about how to measure this. There is a way to go on this and we are open to ideas for sure, if anybody can give us any.

Dr Halpern: With the demise of HOCS, the new Cabinet Office survey, which is quite a lot cheaper, will cover some of those issues. I should confess that I did write a book on social capital a number of years ago, so I am a bit more optimistic about the measurement. If you put in lots of different measures, including voting and connection and so on, and if you had to ask just one question on social capital: "Do you think other people can be trusted?" tells you a lot about your community or your nation and so on, and there are massive variations across countries. In Scandinavian countries, the levels of social trust are such that the answer to that question would be in the range of 60% to 70%, by contrast in South American countries it is less than 10%, generally speaking.

Q56 Caroline Lucas: Where would we be?

Dr Halpern: We are about 30% now. It is predictable for all kinds of things, including the economic growth rate. For example, will people recycle? Do you recycle and pay your taxes if you do not trust anybody else to do the same? It is a very powerful and ubiquitous variable.

Caroline Lucas: Thank you.

Q57 Caroline Lucas: I also have a question on housing. As I understand it, the housing indicator simply measures the annual increase of housing stock, and my question is: would it not be more relevant to measure the level of housing availability in proportion to the level of housing demand, so it is a more relative figure to take into account population growth or growth in the number of single households requiring housing?

Nigel Atkinson: You are right. The indicator we have adopted is a measure of supply, not a measure of demand. Creating a measure of demand, however, is not straightforward. Where that demand is effective, that is to say it is backed by purchasing power or ability to rent or whatever, then it is essentially reflected in market transactions and therefore implicitly partly in this indicator. The problem is measuring the extent of demand that is not being met given the current prices, mortgage availability, rents, and so on. I guess one would also want to be aware of the geographical dimension to that problem. What we have put in the contextual indicators is a measure of household projections-that is the indicator in the supplementary set-so essentially what that gives you is projected growth in the number of single person households, or two person households, and so on, through another 10 or 20 years. That is probably the closest thing we have to a measure of demand, which is why we are suggesting that the headline measure we have proposed is read against that contextual indicator. We did look at other things, for example we considered looking at the provision of affordable housing-indeed that is a business plan measure for CLG-but it is actually still a measure of supply rather than demand. For the purposes of an overarching indicator it seems to us to perhaps be a bit narrow. That is the reason we have gone the way we have.

Q58 Caroline Lucas: A cheeky question, but was it not driven at all by the fact that Ministers might quite like to have a focus on additional house provision and looking at it in terms of additional house building as a discrete policy objective?

Nigel Atkinson: No, we suggested this.

Q59 Chair: The new natural resources SDI monitors consumption of particular classes of resources. I am a little bit confused as to why it includes bio-mass and timber, which would presumably be seen as sustainable crops, and why fossil fuels, and their consumption, is measured against GDP growth. There does not seem to be a consistency in terms of resource efficiency policy.

Nigel Atkinson: I think you are right; the measure we have adopted combines renewables and non-renewables in one indicator.

Q60 Chair: Is there any reason for that?

Nigel Atkinson: The reason is purely simplicity, but we recognise that there is a case for looking at these things separately for the reasons you have mentioned. It does seem that we ought to be able to put together a data set that does that.

Q61 Chair: But you are not doing that, are you?

Nigel Atkinson: It is not what we proposed in the consultation, but I think it is one of the things that we will want to look at again in the light of comments we have received and your views as well. It does look as though it ought to be possible to do that, so we will look at it. The other thing that we might look at in that indicator is the fact that we excluded aggregates from the indicator, because this is a measure combined on the basis of physical volume that skewed the indicator hugely in one direction. On reflection, we think that maybe it would be worth showing that separately within the context of the indicator, so we will think about that also.

Q62 Chair: Just staying on that theme of resource efficiency. As I understand it, the indicator does not take account of where production has moved offshore and where you have embedded carbon that is coming in; is there not a case to have greater clarity about the industrial base that we have in this country?

Nigel Atkinson: That is right in part. What the indicator does is look at the natural resource content of our consumption. In doing that, it estimates the natural resource element of our imports. If our economy shifts from manufacturing to services such that we end up importing more of our needs in terms of manufactured goods, then it would pick up the immediate raw material consequences of that. So, in that sense it does the job, although of course there are potentially wider implications of all that. What it is not, as you are saying, is a measure of the efficiency of the use of our natural resources. It does not do that and does not attempt to do that.

Q63 Chair: Is it likely to be something that you will be picking up following the close of the consultation?

Nigel Atkinson: We will certainly want to look at it. I imagine we must have looked at this in the past and decided it was rather too difficult to do.

Q64 Chair: Did you look at it in the past?

Stephen Hall: In terms of the question you had about embedded emissions, we do have an indicator on greenhouse gas emissions based on a UK consumption basis. That is taking account of emissions embedded in our imports. As for the measure of natural resources, in our consultation document it was labelled as experimental data, although there are all manner of different ways in which you could combine data to create some measure of natural resource use, and this is just one of them that has been developed recently. It is experimental, we will continue to develop it and, as we have intimated, we will be able to break this down into more detail as we develop the data set.

Q65 Chair: Clearly, the danger is that we could be doing very well in reaching that indicator if we were being resource efficient, but if it just so happened that all the production was out of the UK, it would not be telling very much about how resource efficient we are in terms of the goods we use, as opposed to the goods that we manufacture and produce.

Nigel Atkinson: Yes. It tells us about the volume of raw material that is being consumed implicitly; you are right that it does not tell us anything about the efficiency of production.

Q66 Chair: So would you be looking for some modification of that?

Nigel Atkinson: I certainly agree that it would be very useful if we could do that.

Q67 Chair: Are you already looking at that, or will you be doing it from now onwards?

Stephen Hall: I am not an expert on this particular indicator, but I think this may be taking into account the resources that we are taking from outside the UK in order to meet our consumption requirements. We can clarify this afterwards.1

Chair: That is helpful, thank you.

Stephen Hall: As I said, there are many different measures, some of which do exactly what you are asking.

Q68 Chair: In response to the question we have just had from Caroline Lucas, you mentioned about resources not necessarily being there to do some of the work on some of the indicators because they are not yet properly developed. Are you confident that you have the resources to do the outstanding work on those indicators that have not yet been fully worked up?

Nigel Atkinson: In all of these cases, we are trying to make a judgment about the relationship between benefits and costs. The development of some indicators would look to be a much more expensive matter than others, inherently because of the data problems involved. In each case we are trying to make a judgment in advance about what the benefits would be of working on producing that indicator relative to another one. Given the almost limitless possibilities here, we are always going to be in that work.

Q69 Chair: I certainly think that this Committee would be very interested to know which ones, and the cost of doing so might not justify the necessary outcome. If you could perhaps give us an indication of where some of them may fall by the wayside due to lack of justification in terms of cost, that might be helpful.

Nigel Atkinson: Okay. Also, when we look at this and we look at the potential benefits from working on one indicator relative to another, one of the things we want to take into account is the demand for that as well. Your views on which are the priorities among these possibilities would also be welcome to us.

Chair: We can look at that.

Stephen Hall: I just want to add that I do not think in any instance the indicators that we have in the set are only there as sustainable development indicators. We have tried to tap into efforts being undertaken across Government to produce measures for other purposes. In all instances the primary customer would not be us, the information would be produced for other purposes and then we would be making best use of existing data across Government.

Q70 Chair: Absolutely finally, in the light of the national infrastructure legislation introduced to Parliament last week, when can we expect the work to be done in respect of land use development linked into the NPPF? How, for example, are you expecting to differentiate between green field and brown field sites? Is that on your agenda?

Nigel Atkinson: We have received approximately 2,500 related responses, and they ask us to look in more detail at indicators of the extent to which urban sprawl is being prevented, in terms of looking at the proportion of all new house building on brown field land and on how many new homes are built on a hectare of land. That is clearly sending us a message and that is something that we will therefore look at in response to consultation.

Chair: Thank you to all four of you for your time this afternoon, and we will see where this agenda leads to. Thank you very much indeed.