Joint Committee on Finance, Public Expenditure and Reform, and Taoiseach debate -
Thursday, 13 Oct 2016

National Economic Output: Director General, Central Statistics Office

Chairman

In addition to discussing the CSO's reservations on the EU legislative proposals outlined above, the joint committee agreed to engage with the CSO on the methodology it employs for calculating national economic output. I welcome Mr. Pádraig Dalton, Director General of the CSO, to today's meeting to discuss these topics. Before we begin, I draw the attention of witnesses to the fact that by virtue of section 17(2)(l) of the Defamation Act 2009, witnesses are protected by absolute privilege in respect of their evidence to the committee. However, if they are directed by the committee to cease giving evidence on a particular matter and they continue to so do, they are entitled thereafter only to a qualified privilege in respect of their evidence. They are directed that only evidence connected with the subject matter of these proceedings is to be given and they are asked to respect the parliamentary practice to the effect that, where possible, they should not criticise or make charges against any person, persons or entity by name or in such a way as to make him, her or it identifiable.

Members are reminded of the long-standing parliamentary practice to the effect that they should not comment on, criticise or make charges against a person outside the House or an official either by name or in such a way as to make him or her identifiable.

I invite Mr. Dalton to make his opening statement to the committee.

Mr. Pádraig Dalton

I thank the Chairman and the committee for inviting the Central Statistics Office here today. I am accompanied by my colleagues Ms Jennifer Banim, Mr. Richard McMahon, Mr. Michael Connolly, Mr. Brian Ring and Mr. Ciaran Dooly. The CSO operates under the terms of the Statistics Act 1993, which sets out the mandate of the office and the standards by which we must conduct our business. Independent, objective, trusted and high quality official statistics are the cornerstone of any developed democratic society. A key element in ensuring trust and quality is the legal guarantee in relation to the confidentiality of statistical returns made to the CSO. Accordingly, we cannot discuss any issues relating to any individual person or entity. Due to the nature of our work, the CSO is entrusted with a large quantity of extremely sensitive information by our respondents, both businesses and persons. The legal guarantee we provide to these respondents is central to our ability to collect information and to the compilation of all the official statistics we publish.

The CSO operates in a dynamic global environment, driven not only by a shift in focus towards greater levels of scrutiny but also by the unprecedented growth in the volume and nature of data. The CSO embraces the opportunities for official statistics provided by these changing trends. We are engaged in a programme of modernisation and innovation and we are respected contributors on the international stage. The CSO is the main producer of official statistics in Ireland but there is also a significant international dimension to our work. Meeting EU legislative requirements, which have a considerable overlap with national requirements, accounts for almost 90% of our work programme and has been the single most important factor shaping the development of the CSO since joining the EU.

We have been invited here today to discuss three particular issues. First, is the proposal for a regulation of the European Parliament and of the Council establishing a framework for European statistics relating to persons and households. This is known as the integrated European social statistics, or IESS, regulation. Second, is a proposal for a regulation of the European Parliament and of the Council on the European statistical programme. Finally, the committee wishes to discuss the methodology for calculating national economic output. I propose to take each of these in turn.

I begin with the IESS regulation. Currently, the production of European statistics on persons and households is governed by a suite of five stand-alone, topic-specific EU regulations. The proposed framework regulation is designed to address this fragmentation and will provide a unified legal basis for the existing collection of social statistics from persons and households in sample surveys while enabling future integration and improvements. The primary objective of this modernisation initiative is to move away from the traditional way of producing social statistics in silos to a more integrated legal framework leading to a standardisation of concepts and greater comparability across member states and surveys. The proposed consolidated regulation will cover the labour market, income and living conditions, health, education and training, the use of information and communication technologies, time use and consumption.

The CSO supports the modernisation initiative and the overall objectives of the proposal. We agree with the consolidation of the legal framework, the integration of data collections, standardisation of variables and wider use of innovative data sources. However, we feel that the priority areas identified do not adequately reflect user needs and are inconsistent with the Beyond GDP initiative which focuses on broader quality of life indicators. The CSO opposes some specific elements of the regulation including the move towards input harmonisation, which prescribes how member states ask and collect information; the overburdening of the survey on income and living conditions; and the legislative vehicles being proposed by EUROSTAT which would significantly reduce member states' control over survey costs and burden.

I turn now to the EU regulation to amend the European statistical programme 2013 to 2017.

The current programme extends to 2017 and provides the framework for the development, production and dissemination of European statistics. The extension by three years to 2020 brings the statistical programme into line with the multiannual financial framework applied by the Commission and takes account of new policy demands.

In addition to the existing outputs of the current programme, this amended programme aims to produce a broader range of information on topics, including inequality, poverty, material deprivation, services sector, energy and climate change. There is also a focus on measuring progress on the UN Sustainable Development Goals.

While the CSO is broadly in favour of the amended statistical programme, we intend to challenge on certain amendments during the co-legislative process. In particular, we are seeking clarity on the exact requirements before agreeing to the inclusion of annual updates of population projections and the development of any additional social survey outside the realm of the integrated European social statistics framework.

Finally, I will turn to the methodology for calculating national economic output. The highly globalised nature of the Irish economy was demonstrated dramatically in July with the publication of the national accounts and balance of payments statistics for 2015. The CSO compiles national accounts and balance of payments statistics in accordance with the standards set by the UN and the IMF and as required under EU legislation. Adhering to the international standards ensures the provision of results that can be compared across countries and across time. The statistics published by the CSO in the 2015 national income and expenditure publication, were compiled in accordance with these standards and accurately reflect the highly globalised and complex nature and structure of the Irish economy. In this context the figures are highly relevant and play an important role in understanding the complexity of our economy. However, as the CSO has previously stated, the headline indicators of GDP and GNP are now of limited value in providing insights into the domestic economy.

The level shift in GDP of 26% in 2015 was driven by relocations of entire balance sheets to Ireland, with the activity related to these relocations having consequential impacts on the results. The relocated balance sheets were dominated by intellectual property categorised as intangible assets. The practice of relocating intellectual property to Ireland has been growing in recent years, but the scale of the relocations in 2015 was substantial and added €300 billion to Ireland’s capital stocks. Associated with the relocations were significant increases in contract manufacturing activity attributable to Ireland. When the net effect of sales of products produced abroad under contract were added to Ireland’s trade in goods, the balance of trade in goods and services in the national accounts doubled from €35 billion to €70 billion between 2014 and 2015.

In the past, the impact of contract manufacturing activities on exports of goods was largely offset by imports of research and development services, as Irish companies made payments to non-resident parts of the group for the use of intellectual property. However, when the intellectual property is located in Ireland, these offsetting charges do not occur, and the full effect of contract manufacturing is attributed to GDP, as seen in the results for 2015.

Additionally, the capital assets in the relocated balance sheets have also led to significant increases in the estimates for depreciation of assets in the national accounts. As a result of this increase in the depreciation charged in Ireland, foreign direct investment profits attributable to the rest of the world are reduced. These profits are a major part of the difference between GDP and GNP. The lower levels of profits, due to the increased depreciation charges, leads to a narrowing of the gap between GDP and GNP. GNP is often seen as a better measure of the underlying level of economic activity in Ireland, but as an indicator, it is now also elevated by these relocations.

To illustrate the impact of the relocations on GDP and GNP results for 2015, net national product, an alternative indicator, also published by the CSO in July from the national accounts framework that measures economic activity after the effects of profits and depreciation, grew by 6.5% between 2014 and 2015, compared to the 26% growth in GDP.

EUROSTAT visited the CSO on 30 August as part of the normal gross national income, GNI, verification process for national accounts and balance of payments statistics. The meeting was very productive and the results for 2015 were verified. EUROSTAT's summary report of the visit will be discussed at the upcoming GNI committee meeting on 17 and 18 October and the minutes of the GNI committee meeting will be publicly available.

Ireland's macroeconomic statistics are built on the information supplied directly by companies and our broader range of data providers to the CSO’s large cases unit, LCU. The LCU is a unit dedicated to engaging with the largest 75 companies in Ireland.

As well as data supplied directly by the multinational enterprises, MNEs, the large cases unit has access to a wide range of data from CSO's balance of payments and business statistics surveys. Additionally, the LCU has access to data sources such as trade in goods data and corporation tax data. The broad range of information available to the LCU ensures consistent recording of activity across CSO outputs and its engagement with the largest multinational companies ensures that the complex structures of multinationals are correctly profiled and reflected in the statistics. In 2015, the companies covered by CSO’s large cases unit accounted for approximately 70% of industrial production, 60% of exports of goods, 40% of exports of services and 60% of imports of services.

The CSO is one of the few national statistical institutes in the EU to have a unit dedicated to profiling and to ensuring the quality and consistency of data provided by large multinational enterprises, MNEs, and the unit has developed significant expertise in globalisation activities during its years of operation. This expertise is recognised internationally through, for example, CSO’s chairing of the United Nations Economic Commission for Europe task force on measuring global production and invitations to CSO to participate in the global inter-secretariat working group on national accounts where changes to the international standards for national accounts statistics are discussed.

In a broader context, the growth in globalisation activities presents a significant measurement challenge for the CSO, EUROSTAT, the UN and the IMF, particularly around providing indicators of domestic performance for national users. It is increasingly difficult to represent the complexities of activity in highly globalised economies, such as Ireland’s, in single headline indicators such as GDP and GNP. The CSO currently publishes sector accounts as part of the national accounts framework which give insight into the household sector and also publishes a breakdown to separate economic activity by multinational enterprise-dominated and indigenous sectors.

A cross-sector group has been convened by the CSO to discuss how best to meet user needs with national users and international observers represented on the group. Indicators currently published by the CSO such as information on personal consumption and data on employment and earnings will form part of the solution, but the discussions of the group will also cover whether there is a need for additional indicators or for developments of detail in existing indicators.

The group has met twice and will meet a further four times before the end of November 2016. The group will report to me, as Director General of the CSO, and the report will be published on the CSO’s website along with the CSO’s response to the report.

I thank the Chairman.

Deputy Pearse Doherty

I wish to raise the concerns Mr. Dalton expressed about what is coming from the European Union. There was a suggestion earlier that some of the information the CSO was requested to compile was quite sensitive, that it would have required information on gender violence and sexual abuse and the CSO believed it was not the appropriate body to collect that type of information on the doorstep and that it could lead to intimidation and retribution. Will Mr. Dalton elaborate on that?

Mr. Pádraig Dalton

In a sense this links both the European statistical programme and IESS. The particular survey which Deputy Doherty referred to was on the table at European level approximately five years ago and it never made its way through the legislative process because there was not enough support across member states. There was a great deal of concern about how one would collect sensitive information from people as young as 16 years in a household. For example, if one were trying to collect information on domestic violence, the perpetrator may also be present at the time. At that stage, there was a certain level of uncertainty in the statistical world as to how best to gather this information.

The current European statistical programme has not currently put that proposal on the table, although it acknowledges the need to collect information in this regard.

We would also agree this is an important area we should cover. The question is what is the best mechanism to collect this type of information to ensure we do not put anybody at risk. We must also ensure we collect accurate information and provide that evidence for policy makers and decision makers to address what is an extremely important issue.

Senator Gerry Horkan took the Chair.

Deputy Pearse Doherty

Does Mr. Pádraig Dalton have a view as to what is the best way to collect this type of information without putting anyone at risk?

Mr. Pádraig Dalton

This is not an area of expertise for us as we have never undertaken a survey of this nature. I met with Rape Crisis Network Ireland at one stage to discuss how one might go about doing this and collaborate nationally with such an entity to help us with the fieldwork.

The discussion at EU level is how one ensures one collects accurate information on this topic which is a sensitive and difficult area. One can imagine going into a household to collect this information, talking about domestic violence when the perpetrator may be there. There is much research going on in this area. Some propose that it is much better to do this through computer-telephone interviewing, rather than doing it the traditional way of calling to the door with a laptop or a palmtop collecting the information face to face. It has been suggested the best way to do it might be with telephone or web-based interviewing.

Deputy Pearse Doherty

Does that type of data collection not form part of the integrated European social statistics regulation at this point?

Mr. Pádraig Dalton

No, not at this point in time.

Deputy Pearse Doherty

Mr. Pádraig Dalton said it is inconsistent with the beyond gross domestic product initiative, which focuses on the broader quality of life indicators. He also stated there is a concern with the overburden of the SILC, survey on income and living conditions, and the legislative vehicles being proposed by EUROSTAT, which would significantly reduce member states' control over survey costs and burden. Can Mr. Dalton break down what that really means for the Central Statistics Office, CSO? What type of additional information would the CSO have to collect which could overburden the SILC? Would the CSO still retain complete autonomy in the collection of data and its analysis? What does this really mean in terms of the harmonisation under this regulation?

Mr. Pádraig Dalton

The first action the integrated European social statistics, IESS, regulation is trying to do is to harmonise the definitions of particular variables across surveys. Currently, one has a regulation for the labour force survey and a separate regulation for the SILC. Taking a simple example, the definition for a "usual resident", the primary basis for the people covered in a survey, differs from that in the labour force survey to the one in the SILC. That creates problems when one is trying to correlate it on an aggregate level between labour market data and data on SILC. One of the key things they are trying to do at European level, which we support, is to ensure standard definitions for these core concepts and topics and that the same definitions apply across all the surveys.

A respondent is in the SILC for four years. There are two parts to the survey. There is the pure cross-sectional part which is where we provide statistics for a given year. Then there is the longitudinal component where one is trying to track changes over time by taking a cohort of people and monitoring how their circumstances change over time. In Ireland, we have three waves for the first three years and then one rotates out at the end of the fourth year. The proposal at European level is to move this to six waves. The problem we have with all surveys of this nature, not just in Ireland, is that one is trying to recruit people into a survey. One is accessing information on sensitive topics such as income, material deprivation and measuring consistent poverty. The first problem we face is recruiting households into a survey. Once they are recruited, like every other country, the attrition rate is quite high because of the sensitive nature of the topics. As one moves through the four years, the coverage of the longitudinal component reduces. Imagine going to a doorstep and asking someone to engage with the survey for six years. The concern we have, as do other member states, is the shift from a four-wave approach to a six-wave approach.

We will retain total autonomy of the SILC. The SILC is a long survey, taking about an hour to go through it. One is trying to engage householders who are busy and it takes place in the evening. The proposal is now to include detailed questions on wealth, on top of an already difficult survey. Our view is that this is not the appropriate place to capture wealth. The appropriate place to capture wealth is in a dedicated survey. This is currently organised by the European Central Bank, ECB. The CSO undertook the first wealth survey, the household finance and consumption survey, with support from the Central Bank in Ireland. That is an incredibly difficult survey, looking for access to information on income, which is current, but also on equities, deposits, house ownership and so forth. It is a very difficult survey. Trying to put wealth on top of a survey on income and living conditions is not feasible or sensible. We are not the only ones who feel that way. We will have to see how this proposal progresses through the co-decision process.

Deputy Pearse Doherty

I must compliment the CSO on the significant amount of information it compiles and presents on its website. Making it as user-friendly as possible is some achievement. We as legislators rely quite heavily on it to get the correct facts and data.

On the revision of the GDP figures of a 26% growth rate this summer, the CSO came under intense criticism, including from Members on the floor of the Dáil. The country was labelled with the tag "leprechaun economics" and the blame was laid at the CSO's feet. What does Mr. Pádraig Dalton say to those who blame the CSO for this revision?

Mr. Pádraig Dalton

The best way to answer that question is to talk through the process.

We compile our national accounts data, based on hard facts provided to us by our respondents. We also use a broad range of additional data sources. The CSO is uniquely placed, not just in Ireland but across Europe, in that it has access to as broad a range of data as anybody can have when compiling national accounts. We compile the data in accordance with the standards. It is important to recognise there is no Irish, UK or French version of gross domestic product or gross national product. There is only one standard. We are legally bound to compile these figures in compliance with that standard. We did that. We knew the publication of these figures was going to create a reaction. We did the right thing in what we knew were going to be difficult circumstances. This is one of the strengths of the CSO, namely our independence, objectivity and integrity for data.

What is very important to recognise is that the accuracy of the data has not been questioned by anyone. The figures were verified by EUROSTAT. The IMF, the UN and the OECD have not questioned the accuracy of the data. Instead, they have all spoken about how the publication of the figures in Ireland this year provided the basis for a conversation which had been going on for years around how one deals meaningfully with globalisation in a national accounts perspective. This had been happening in a conceptual way up until this year. Our publication this year has moved the conversation into the real world.

The most important thing for us is that the figures we compile are accurate, independent and objective.

Deputy Pearse Doherty

I agree. Mr. Dalton mentioned that GDP and GNP are of limited value in providing insights into the domestic economy. The figure of a 26% rise in GDP, published last July, was an accurate reflection of the data in terms of how we calculate GDP. The reality, however, is that GDP, in so far as how it is measured and how people interpret it, no longer provides us with a proper metric. That was not just the case this year; there was evidence of it in previous years. We heard about how contract manufacturing in previous years made up a portion of GDP figures. When did a perception of the limited value of GDP begin to creep in, particularly in the context of globalisation?

The witnesses noted that intellectual properties being categorised as intangible assets and moved on to Ireland's balance sheets created €3 billion of additional capital stock. I presume we are talking about a small number of companies here. If the trend in intangible intellectual property were reversed, Ireland would find itself with a major drop in GDP, which could put us into a statistical recession and see our debt to GDP levels skyrocketing which might, in turn, interfere with how the bond markets and so on deal with us. That is as likely as what we saw last year. How do we safeguard ourselves in a situation where a limited number of companies are moving intellectual properties between different jurisdictions, thereby screwing up our national accounts and potentially having various unforeseen consequences? We are already seeing the contribution we have to pay to the EU increasing as a result of the situation in regard to intellectual property. There could be wider implications if the situation were reversed.

The 26% figure was a revision, with the CSO previously indicating that GDP had risen by 6.7%. I assume that revision was done in consultation with the large cases unit within the office, which deals directly with the 75 companies. How did it transpire that the data provided by some of those companies to the large cases unit was incorrect and had been amended? Was it the case that those firms did not provide accurate and timely information to the CSO, leading to the original indication of a GDP increase of 6.7%, and came to the office later saying they forgot to mention they had moved all their intellectual property from California to Cork or wherever? Was the CSO left to deal with the fact there was an additional €3 billion that had to factor into the accounts? Is there any requirement on any of the large-case companies to provide the CSO with specific types of information and, if so, are there any penalties for their failing to do so? Is the CSO following up any such failures?

Mr. Pádraig Dalton

The Deputy asked when concerns about the usefulness of GDP and GNP first crept in. The debate on that issue in this country has been going on for many years, since the 1980s. At that time, the globalisation activity everybody was talking about was profit repatriation and transfer pricing. Things have moved on considerably since then and now we are talking about contract manufacturing, corporate inversions, relocation of assets and so on. The concern about GDP and GNP is not new; in fact, it goes back right to the 1940s. Simon Kuznets, who developed the metrics, was of the view that post the recession of the 1930s, the focus in measuring economic activity should shift back to an income measure rather than a production and output measure. What has happened since then is that from time to time, events will tend to resurface the debate. Certainly, the publication of our figures last July resurfaced the debate in this country in a very real way.

We fully understand the difficulties the publication last summer of the GDP and GNP figures caused for our users. We are not immune to that and are now trying to move the narrative beyond GDP and GNP. That is difficult, however, because - not just in Ireland but globally - those metrics are the denominators for everything. It is a conversation that needs to take place beyond the statistical community. The legislative and policy community must also consider whether, in an environment where economic activity is increasingly globalised, GDP and GNP are appropriate sources of information. We have already surfaced this discussion and raised the issue at European level within the statistical community. Current practice will not be changed by statisticians, however, because we measure what we are told to measure. The debate must happen elsewhere.

Going back to 12 July, the national accounts framework contained a great deal of information other than the GDP figure. For example, the net national product measurement, which we published on the same day, showed a growth rate of 6.4%. We published a figure for personal consumption and expenditure, which was showing a growth rate of 4.4%. Employment, the data showed, was up 2.3%. Last week we published sectoral data which indicated a growth rate for the indigenous sector of 4.4%. Household income data were published the week before last from a national accounts perspective, showing growth of some 5.5%. It is perfectly understandable that those figures were lost in the haze of the growth figure of 26%. We in the CSO have a job to do to give greater prominence to a broader range of issues. However, if we move in that direction while the EU and international community continue to base everything on GDP and GNP, it will be a very difficult battle. We have already commenced discussion on these matters with international partners, including the UN. My colleague, Ms Banim, has been asked to travel to Washington in November to deliver a number of talks and papers about what is happening in Ireland. People want to learn from our experience because they know it is an issue. Indeed, how Ireland has dealt with these matters is seen as a case study of how other countries might proceed, while pointing up the significant issues that may arise.

The Deputy's second question was whether the movement in intellectual property could go in the opposite direction, to which the answer is "Yes, it could". There is no doubt it could happen. Our view, however, is that as long as the environment persists which has given rise to the trend whereby IP has, for some years now, been locating in Ireland, we are unlikely to see a shift the other way. In fact, we in the CSO expect we could actually see additional level shifts - not to the scale we saw in July but certainly additional level shifts in our GDP and GNP, because that environment still exists. As such, there will continue to be question marks over GDP and GNP, which gives an urgency to the need for all of us to start thinking beyond those metrics and looking at other indicators like net national product, personal consumption and expenditure, the data contained in the institutional sector accounts we provide, and the indigenous sector figures we publish.

On the revisions, the first estimate of 6.7% was based on the suite of quarterly returns up to the end of 2015. It is only at the end of the year, when all companies, particularly the multinational enterprises, are finalising their accounts, that they make decisions around pricing and location of certain issues. That is why we end up having to revise the four previous quarters, because the multinationals, especially, make changes at the end of the year. Early in 2016, when those entities were finalising their annual accounts, our large cases unit became aware of a number of cases that raised our antennae and led us to undertake a series of direct site visits. In some cases, we had up to nine visits with the companies in question over a period of four to five months. It is a challenging task because each of those companies might have 40, 50 or even 60 subsidiaries or affiliates. Our first job, then, is to profile the entity and find out how many affiliates it has. Second, we need to ascertain its residency for statistical purposes. Finally, we must discover the balance sheet of the entity, its assets and liabilities. That was a block of work we had to do over the first four months of this year.

That is why we have moved from this figure of 6.7%.

It must be remembered that the change in structure in some of these entities actually took place in the first quarter of 2015. As soon as we found out about it, we engaged directly with the companies involved and undertook the site visits. I think this is a particular benefit of our large cases unit. The large cases unit in Ireland is seen as best practice across Europe. Since July, a number of national statistical institutes have been in contact with me wanting to come and visit the CSO to see how it operates. That is how it happened. I hope I have answered the Deputy's question.

Deputy Pearse Doherty

The last point the witness makes is quite interesting about it being the first quarter in 2015 when this change took place. That is what the witness mentioned, yet we had the revision quite a bit later. It is interesting that the CSO's antennae was raised. Would it be helpful if there was a requirement on some of these large multinationals to inform the CSO, as opposed to it having to have its antennae raised for something that is to have a dramatic consequence for the country?

A new metric committee is being set up. The Central Bank and the CSO will be involved and the Central Bank will report to the CSO. However, the new metric will have no legal basis. It may help us to have a discussion about how our country is going, how the economy is growing, is it really growing, is it in recession and so on. We will be able to have a more informed debate. The problem for the State at this point in time is that we have legally-binding targets that are measured not by whatever the new metric committee will come up with, though I wish it well in terms of its work, but against GDP. With all of the conversations going on in the UN and the EU, it is unlikely that will be changed in the short term. I am not sure whether there was any other country in the world that had a 26% increase in GDP last year. We have an economy that is very reliant on multinationals, has allowed for a lot of corporate inversions, contract manufacturing and quite unique activities, though they happen in some shape or form elsewhere, but that are very concentrated in this country and that have thrown off our figures. It has basically meant that they are meaningless for us as a metric. What are we going to do?

The point I made to the witness is that if one of these companies decides to move its intellectual property to America next year, we could be looking at a GDP reduction of 20%. Then there is the issue whereby we have legally-binding targets for debt-to-GDP to be reduced or deficit-to-GDP to be reduced. We would have a real problem and there could be international spillovers in terms of the international bond markets. That would cost all of us. Is there any interim solution? I know that the long-term solution is to have a more appropriate measure internationally. I agree with the witness on that. However, whatever we come up with will be an Irish solution to Irish problem and will not be recognised by anybody else in the world. Has the witness any interim solution to try to deal with this? Is there a way of dealing with this?

Mr. Pádraig Dalton

I think there is and I will outline that in a second. One of the things that I want to pick up on is something that we are maybe a little bit conscious of. I fully understand the broad statement that our GDP and GNP rates are meaningless. However, in a sense, they are not meaningless when one thinks about the debate we have been having since July. We have had an incredibly informed and considered debate in this country since July on a very important issue. While GDP and GNP are not useful indicators in regard to the performance of the domestic economy, what they do tell us about is the structure of the Irish economy as a whole. What they have highlighted to us is the sheer scale of globalisation activity. We all knew that we had a globalised economy. We all knew that we had a quite concentrated economy in which a relatively small number of large multinational enterprises dominated our economic figures. However, I do not think that any of us fully appreciated that the impact could have been this large. In that context alone, I believe the figures are very meaningful and have highlighted an important issue.

I believe the Deputy is right in that there is a broader issue here and we need to address it. What we do not need is an Irish solution to an Irish problem, though there will probably have to be an element of an Irish solution to an Irish problem. We need to think about the broader national accounts framework. There is a misconception that the national accounts framework is just about GDP and GNP. It is not. There is net national product, net domestic product, personal consumption and expenditure, and national accounts for the indigenous sector, which the CSO has done, vis-à-vis the multinational enterprise sector. We can produce the institutional sector accounts and the figure I quoted earlier was about the households. In a sense, the household figure is how Joe Public on the street feels the economy is performing. That gives one a sense at an aggregate level of what is happening to him and how things are performing for him. There are indicators that are already there that would be very useful. We might have to improve the frequency. We only produce some of them annually and I think we might have to do some work to start producing them quarterly. Some additional breakdowns of the data would be very useful as well. They are all the things that we are currently working on. We are not trying to pre-empt the work of the expert group. I am sure they will come forward with some additional indicators. We have already started thinking about this ourselves.

The big issue the Deputy mentioned, and the one that I do not have an answer to is at an EU level and everything being based on GDP and GNP. I suppose that, as statisticians, we cannot change the law. That is not the realm for statisticians. We can only deal with the issues that we are asked to measure. The decision around the deficit being measured as a proportion of GDP, or the debt being a proportion of GDP, those decisions are taken beyond the statistical realm and that is a debate that needs to take place at a policy level. It is an issue for legislators to look at at a European level. Whether that can change quickly, we do not know. We can only talk about the statistical side of it.

Senator Kieran O'Donnell

I thank Mr. Dalton. I have a couple of questions. How many multinational companies were involved in that revision?

Mr. Pádraig Dalton

Just to keep it simple in one sense, we place a legal obligation on companies to provide information to us. In doing that, we provide a legal guarantee to them that we protect their identity and the data they provide to us. All I can say is that there was a small number of multinational enterprises involved.

Senator Kieran O'Donnell

Less than five?

Mr. Pádraig Dalton

We cannot go beyond what we have said about a small number. I want to contextualise it a little bit. Our ability to compile official statistics and to provide evidence for decision-makers like the members and for decision-makers outside of the policy of Government is very important. If we breach the trust that respondents place in us by divulging any information, either directly or indirectly by anything we could say, we will not be in a position to compile economic statistics. As far as I can go, I am not going to say it was less than five, ten or 15. All I can say to the Deputy is that it was a small number of enterprises and I cannot go any further than that.

Senator Kieran O'Donnell

In the limited time, how many people are employed and working in the large cases unit in the CSO?

Mr. Pádraig Dalton

I will ask my colleague, Mr. Michael Connolly, who is the head of the large cases unit, to answer that.

Mr. Michael Connolly

There are eight full-time staff, but the Deputy has to understand that there is a lot of additional support that comes from our business register in Cork.

Senator Kieran O'Donnell

What are they qualified as?

Mr. Michael Connolly

They are a combination of professional accountants, financial economists, statisticians and then-----

Senator Kieran O'Donnell

Does the witness think it is big enough to deal with-----

Mr. Michael Connolly

I think it is probably never big enough, though one could always say that. We certainly have enough time to deal with these major companies in a consistent way. As Mr. Dalton was saying earlier, our objective is to get a consistent view across all of the various survey-type data and then the administrative data that comes to our international trade, our revenue data and so on. It gives us this comprehensive understanding of their activities.

Senator Kieran O'Donnell

Does Mr. Dalton believe that this change in 2015 was a direct result of the abolition of the double Irish?

Mr. Pádraig Dalton

Our job is to present the facts, collect the information from the multinational enterprises, compile it to the standards and publish it. We do not collect any information on the motivation for the decisions of multinational enterprises. We would have no information available to us that would allow me to give the Senator an objective and informed answer on that particular question.

Senator Kieran O'Donnell

Does the witness believe this to be a once-off?

Mr. Pádraig Dalton

I would say it is unlikely that we will see another 26% increase. However, as I think I mentioned earlier on, the relocation of IP did not just happen in 2015. This has been a trend that we have seen in Ireland for very many years. The difference in 2015 was the absolute scale of it. I might eat my words and I had better be careful because I am on record, but it is unlikely that we will see a change of that scale again.

It is possible that we will see another of what we would call a level shift in the level of GDP or GNP, as long as the environment that gave rise to the current level shift continues to pertain in Ireland.

Senator Kieran O'Donnell

For 2015 the CSO produced preliminary figures and the big change came when it produced the final figures. That was on the basis of interaction with multinational companies. The ordinary person looking in will ask why that happened. Did the CSO not consider preparing a publication which stripped out the exceptional items and, almost like a structural deficit, gave a structural GDP-GNP figure and then a real figure which would strip out the exceptional items? Did the CSO consider doing that and will it consider doing it into the future?

Mr. Pádraig Dalton

Earlier on, I pointed to the net national product figure, which excludes the impact of depreciation and, therefore, nets out some of the effect that we-----

Senator Kieran O'Donnell

Is it not fair to say that net national product is effectively the GNP figure minus depreciation? Therefore, it is just taking out depreciation.

Mr. Pádraig Dalton

Yes.

Senator Kieran O'Donnell

This means that depreciation must have been a heck of a figure for the CSO to pull it back to where, effectively, the growth rate in net national product was so low.

Mr. Pádraig Dalton

I am going to pass over to Ms Jennifer Banim shortly to give the Senator the details. With regard to net national product, we also publish personal consumption and expenditure and we-----

Senator Kieran O'Donnell

I have limited time and this is extremely important. The bottom line is that, internationally, we are judged like any other country by the markets and the rating agencies based on the metrics of GDP and GNP. We can talk about the broader question but, for the next number of years, which are critical years for us, we are going to be judged on those two key components. If we have another year where the figure is skewed, it is going to have an impact. We are coming into a very difficult period, with Brexit happening next March, so we have a two-year period of major uncertainty. To the ordinary person looking in, ultimately, this will be around GNP and GDP. What is the CSO going to do, as the independent reporting agency on the national statistics, to ensure that if these exceptional items arise, there will be a presentation to the markets and the rating agencies which explains that these are exceptional items and explains what the norm for Ireland is?

Mr. Pádraig Dalton

That is why we have established the expert group, which is going to look at a broad range of issues. It will look at the existing suite of indicators and determine which of those indicators we could give greater prominence to and whether we could provide greater detail on those. It is also going to look at the potential for new indicators that will provide a better insight into what is happening in the domestic economy. We are already working on this and I have already identified some of the work we are doing. The Senator can rest assured we are working very hard on producing a broader suite of indicators that will provide the insights international users need. We have brought together this very expert group of people who will definitely provide some very useful recommendations to us as to the appropriate way for us to move forward.

Senator Kieran O'Donnell

The CSO would have seen the extreme variance between the preliminary figure of 6.7% and the final figure of 26%. As an independent body, knowing how the markets would react, why did the CSO not give the detail that would allow them to strip out the exceptional items of intellectual property and so on? Did the CSO not present two figures, one being the actual figure and the other being the real, ongoing figure with the exceptional items stripped out?

Mr. Pádraig Dalton

I do not want to repeat myself but I think we have done a lot already to put out additional indicators. As I said, we published a lot more than GDP and GNP in July. However, it was understandable that people focused just on GDP and GNP. Given the scale of the change, it was difficult for users to look at the broader suite of indicators that we did provide and that did give that type of analysis. I will ask Ms Banim to continue.

Ms Jennifer Banim

To confirm this for the Senator, we will be continuing to compile and publish GDP and GNP, as we have to under EU legislation. As the Senator said, they will remain the headline indicators that global and national users will look to.

The Senator asked a good question about depreciation and the amount involved to move from GNP to net national product, NNP. As we have outlined, the extent of the relocation of capital assets meant a €300 billion jump in Ireland's capital stock due to the intangible assets that moved at the beginning of 2015. This led to a significant increase in the estimate for depreciation in the accounts, as the Senator has identified.

Senator Kieran O'Donnell

How much of an increase?

Ms Jennifer Banim

I do not have the figure with me but we can get that and pass it on. It was significant, given the €300 billion increase in the assets, and Mr. Connolly might have that information for us later.

We then saw the impact on intellectual property and consequential contract manufacturing on GDP. We saw the impact of the depreciation on GNP because the profits that were being earned here in Ireland by the multinational enterprises were reduced by the large depreciation charge, so there was an impact on GDP and GNP. If one moves to the purest measure we have of the Irish economy currently, given its structure, which is net national product, that removes the effect of profits and depreciation. Therefore, we are getting to something that is within the national account framework that is a legally required indicator. If we could get the discussion focused on net national product, it would move on the discussion. Obviously, it would be a better reflection, when people are talking about the Irish economy, if we can get them to use that.

The other items mentioned by the Senator come back to volatility and what can happen in the statistics. As Mr. Dalton outlined, these things can happen and they can happen in the opposite direction. We have seen increasing globalisation activity attributed to the Irish macroeconomic statistics in recent years and our anticipation is that, given what is happening with Brexit, which the Senator highlighted, it is likely this amount of change and complexity in the national accounts is going to increase and develop. That is why the work of the expert group and the international statistical community to move beyond GDP, if I may use a cliché, is critical.

I would add that, as part of the block of work we are doing in the CSO, the expert group is attended by international observers, so we have people from EUROSTAT and the IMF, and the UN is also interested in contributing to this work. There is an awareness among the international statistical community and the broader community who set out the legislation for what needs to be measured in an economy, that the discussion has to be had. In that sense, I am hopeful we are now serious about a discussion beyond GDP. I think that can feed into something that will really help the Irish user into the future.

Deputy Michael McGrath

I welcome Mr. Dalton and his colleagues from the CSO. Like Deputy Pearse Doherty, I commend the organisation on the work it does. The information it provides is an invaluable source and resource for all of us, and I use its website regularly. I thank the CSO for all of the work being done.

To start with regulation COM 551, does the CSO envisage this will have an impact on, for example, the quarterly national household survey in Ireland, which is the benchmark survey in terms of household measures? I want to raise a concern in regard to surveys generally and imposing greater requirements on those being surveyed. The CSO relies on the goodwill of participants. We have done that survey in my own household so I know it is an imposition and takes time, given it is not just one visit but a series of visits. My point is that the more the CSO adds to it, the more that goodwill will be eroded and it will become hard to get people to participate. Will that survey change as a result of either of these regulations?

Mr. Pádraig Dalton

Broadly speaking, the IESS regulation will bring the quarterly national household survey, the QNHS, or labour force survey - at a European level it is the labour force survey - in under the broader framework. There are no significant changes proposed to the labour force survey. The biggest challenges for us around the labour force survey, and around all household surveys, is exactly what the Deputy has referred to, which is the risk that at European level we decide to overburden these surveys. The CSO is continuously engaging at a European level in always trying to convey the message that when we are engaging with people or with households, it must be acknowledged that people are very busy now. Trying to maintain high levels of response rates is perhaps the biggest challenge for us now in the labour force survey and in all our household surveys. That is the biggest issue. My colleagues, Mr. McMahon and Mr. Ring, manage the labour force survey.

Mr. Richard McMahon

I would like to elaborate on Mr. Dalton's statement. EUROSTAT's proposals effectively ring-fence the QNHS with regard to burden. EUROSTAT is very conscious of increasing the burden but our real worry is with the SILC survey, which is the survey on income and living conditions. The plan is to use that survey as the catch-all survey to measure additional items such as wealth, health and other characteristics. On the QNHS front, I think we are okay from a burden perspective. From the IESS regulation perspective, EUROSTAT is looking at assisting countries in modernising the data collection procedures and will provide support to all member states in moving to computer assisted web interviewing and other data collection techniques.

Deputy Michael McGrath

If it is decided to add questions around wealth, for example, we would see a big drop-off in the compliance or the co-operation rates. That is just a reality of life, despite the respect there is for the CSO and the reassurance it gives in terms of confidence reality and so forth. I wish the CSO well in its efforts to engage in the process and, hopefully, there will be a sensible outcome in the end.

I will now turn to the issue of the national accounts and the GDP revision. In replies to parliamentary questions, the Minister for Finance, Deputy Noonan, referred to one further element of that, which I do not believe is covered in Mr. Dalton's opening remarks. The Minister referred to an increase in new aircraft imports to Ireland for international leasing activities that generate substantial fee income without significant employment effects. The Minister put this in alongside the other measures dealing with contract manufacturing and IP assets being transferred. Will the witnesses explain if that was a significant element also or was it a much more minor element of the revision?

Deputy John McGuinness resumed the Chair.

Mr. Pádraig Dalton

There are four primary globalisation activities that can give rise to changes of this nature, namely, aircraft leasing, contract manufacturing, relocation of assets and corporate inversions. At the press conference in July, the CSO made it clear that aircraft leasing was an element but it was not the significant element that led to the change. As we have said in our opening statement today, the big factor driving the change in July was the combination of relocation of IP and the associated contract manufacturing activity that goes with that.

Deputy Michael McGrath

Sure. In the previous month, June, the Department of Finance published its summer economic statement and predicted real GDP growth of 7.8%, at that stage. Given that the CSO had been working behind the scenes for a number of months, is the CSO prohibited from sending any signal or providing any information, albeit in preliminary form, to the Department of Finance to say there is a major change coming down the track?

Mr. Pádraig Dalton

The short answer is "yes".

Deputy Michael McGrath

Okay.

Mr. Pádraig Dalton

We have a pre-release access policy, which is on our website. Pre-release access is allowed under the European statistics code of practice but it is only allowed in a restricted number of cases. For the national income and expenditure release, there are two named individuals who get pre-release access, one from the Department of Finance and one from the Central Bank. They get the access at 8.30 a.m. on the morning of the release.

With regard to the rationale, if we go back to the core values of official statistics, and I am sure the Deputy is familiar with this, it is really important that when the CSO publishes statistics, the public trusts that what it is getting is an independent and objective assessment of the information. Anything that might undermine that perception would potentially risk that trust. The CSO, like all other national statistics institutes, applies the European code of practice on pre-release access. The direction of travel at a European level is to actually eliminate pre-release access, but it has not gone there now. The most important thing is that there is transparency in who has access and when. The CSO has a register of pre-release access publicly available on its website and it includes the names of the individuals who have pre-release access for those releases, where pre-release access is provided.

Deputy Michael McGrath

So the CSO is simply not allowed under the code, and that is quite clear cut. I think it is important to have that on the record.

Mr. Pádraig Dalton

Yes.

Deputy Michael McGrath

Will Mr. Dalton explain the core objective of the cross-sector group established by the CSO? Is it to come up with a new measure on the domestic economy? Is the measure perhaps to be embraced further afield and be a catalyst for changes to measures internationally? The CSO is already publishing a whole host of information which provides a more meaningful measure of the domestic economy, including retail sales, domestic demands and so forth. What is the core objective?

Mr. Pádraig Dalton

The key block of work will be to identify indicators from within the existing national accounts framework to which greater prominence can be given when we are publishing the national accounts data. The group will also consider if there is additional analysis or breakdown that we can give of those indicators. There is also potential for new indicators. The group will certainly be looking at the potential to develop new indicators that could provide much better insights. The overall focus is all about the insights into what is happening in the domestic economy. The group is also going to make some suggestions about new presentations of existing information that can improve understanding. It is about the CSO moving to providing insight so it is not just about publishing statistics. It is about the idea of moving from producing data and statistics into information and knowledge. I think the group will make some recommendations for us as to how we might improve that. I believe we have already made some improvements. I do not know if the committee has seen the second quarter publication where the CSO produced a lot more than it normally does. We produced the electronic release, a press statement and a very useful question and answers document which answered some of the questions people wanted answered. The Deputy might be aware of the newly launched residential property price index. It had the addition of a video where we talked through what the new residential property price index could do. Next week we will produce a new app where a person can key in a three digit code to get information on the average price of houses in their area. The CSO is continuously improving its communications and dissemination and the group will help also with that.

Deputy Michael McGrath

Who is the authority on the definition of GDP and the methodology for measuring it - the UN, the IMF or EUROSTAT - and what is the legal basis of it?

Mr. Pádraig Dalton

Within Europe the owners would be the Commission which is EUROSTAT, and the legal basis is ESA 2010, the European system of regional and national accounts. ESA 2010 is a derivative of the system of national accounts 2008, which is the global standard. The responsibility and lead for that is the United Nations. Of course, this is not just the United Nations on its own; obviously, there is a huge level of stakeholder engagement with economists and policy-makers to provide input to that. An inter-agency group has been established to start talking about the next set of global indicators. That will take many years to develop. Ireland has been invited onto that group. The CSO's Mr. Michael Connolly has been personally invited onto that group because of Ireland's expertise in globalisation. On a global level, it is led by the United Nations and at European level it is EUROSTAT.

Deputy Michael McGrath

Mr. Dalton made reference to the residential property price index. One of the key measures in the budget, as he knows, is the income tax rebate for first-time buyers. There are some concerns that this would lead to an increase in the price of new houses. Would the information on movements in new residential properties, apartments and houses over the next number of months be easily discernible?

Mr. Pádraig Dalton

The CSO produces the index on a monthly basis and there is a significant amount of information. We can actually break down the profile of the houses and whether they are attached, detached, apartments or otherwise.

I think there is an indicator as to whether the purchaser is a first-time buyer, because the source for this information is the stamp duty data from the Revenue Commissioners. That information would be part of that but I will revert to the Deputy on the matter.

Deputy Michael McGrath

Can it be categorised by new or second hand property?

Mr. Pádraig Dalton

Yes, it can.

Deputy Michael McGrath

Does that apply to the existing data?

Mr. Pádraig Dalton

I am almost 100% on that, but we will come back to the Deputy with a definitive answer. I am almost certain it can.

Deputy Michael McGrath

I thank the Chairman.

Senator Rose Conway-Walsh

I thank Mr. Dalton for his very informative presentation. I wish to discuss the European collection instruction, particularly around the gathering of data on gender based violence. I would have major concerns about that and I am glad that the CSO consulted the Rape Crisis Centre. I suggest that if there are any moves in that direction, there are very useful statistics gathered by Safe Ireland and that anything that is done in the area should be done through Safe Ireland and Women's Aid in order to protect women. Having worked in the area I can see that one needs to be cognisant of a great many safety issues.

The CSO identified a level shift in GDP of 26% in 2015. I certainly think the CSO has done this country and others globally an enormous service in exposing what it did. I know the discomfort around the publication of such a figure because it would hold us up for ridicule in terms of what was being produced. It was very important that the CSO stuck to its guns in terms of the integrity of the findings and exposing that. It is useful information in terms of letting us know how we are exposed as a country in terms of the 75 multinational companies and the concentration of and over-reliance on the activities of multinational companies that can move. In the context of Brexit and related issues, this information is invaluable.

We have discussed the metric, and I have thought for a long time about measurement in the European context and how useful is information on GDP and even GNP in terms of our measurements. I am glad that our statistics are feeding into the global statistics. In Mr. Dalton's opinion, how long will it take before we have a metric that will have a legal basis? I know the level shift of 26% in GDP cost us more than €200 million. I wonder what it might cost us in the future if the shift was the other way round and we were to be exposed to the bond markets? I think that might cost us a great deal more than €200 million. Are we talking about months and years before we come up with a metric? Can we come up with a metric that is solely on a European level or does it have to be a global metric?

Mr. Pádraig Dalton

I thank Senator Conway-Walsh. On the question of how long it will take to come up with a metric, that is difficult to answer because we do not control the process. The precursor to integrated European social statistics, IESS 2010, was IESS 1995. That might give the Senator a sense of how long it can take to change global standards. There is certainly a new urgency that we are consciously aware of at European level. We have been actively engaged since July with the IMF, the UN and EUROSTAT and they are incredibly exercised by the importance of ascertaining how we can address this at European and global level. That is reflected in the fact that we have observers from both EUROSTAT and the IMF on the expert group. The next set of international standards may not be as far away as we think. I think what the Senator is talking about in regard to international standards is definitely years away.

We are talking about multinational enterprises a great deal and that is wholly appropriate. Multinational enterprises play an important role but I would not like us to forget the role of small and medium enterprises in this country. The SME sector is very important from a labour market perspective and in particular in respect of regional employment. The multinational enterprises dominate the level of GDP and GNP, and they are also extremely important employers. Sometimes the SMEs have got lost in the conversation but we put a focus on the SME sector which plays an important role in employment, in particular regional employment.

I think I have covered the main issues.

Senator Rose Conway-Walsh

I think Mr. Dalton is right in terms of small and indigenous industries, but one thing the shift of 26% in GDP tells us, is that policy-wise we need to focus more on adding value to those businesses and ensuring they are here to stay.

I have really wanted to ask this question since 2013. I want to ask about the role of the CSO in employment figures. I am concerned because in 2013, the CSO reported that 29.8% of the 61,000 purported new jobs created that year, that is 26,800 jobs, came from agriculture, forestry and fishing. That figure of 61,000 new jobs was repeated ad nauseam as being an accurate figure when we know there were not 515 jobs created in the agricultural sector each week or 73 jobs created each day at the height of the recession. I know that the CSO entered a caveat in terms of how it was reporting these figures. I think this is where the weakness is but the media and Government ran with the figure that 26,800 new jobs were created in farming, when we knew that was not true. In some way that is reflected in the lack of confidence of people on the ground in the CSO figures. The people on the ground may not read the caveat or understand the qualification of the information. How do we avoid that happening again?

Mr. Pádraig Dalton

I would like to think that people trust the data that the CSO provides. We put a lot of effort and time into ensuring that the information we put into the public domain is accurate and to the highest standard possible. I hope that is the position. Mr. Brian Ring is best placed to answer the question on the specific sector.

Mr. Brian Ring

I thank Senator Conway-Walsh for her questions. I have been part of the digest team since 2009, so I was involved in the statistics for 2013.

The Senator is absolutely correct that we would have informed users of the caveats around agricultural employment. We would have urged caution in the interpretations of the trends at that time. In terms of the message behind what we were saying, we would have acknowledged that we had issued with our figures for agriculture but what we had said at the time and what we still believe is that the sum of the parts was correct, that the overall change in employment at that time was correct but the distribution across what we referred to as the NACE or the industrial sectors at the time was where we had the problem. At the time that would have arisen from the work we have done, which we are continuing to do at present. We are still working towards retrospectively revising the data to take account of what the Senator is saying. We are capturing data from houses which would be termed agricultural. Agriculture would stand out uniquely among the 14 broad sectors from which we produce data as being very focused on the location of the person in employment because agriculture would be very closely linked to farmers and farmers are very closely linked to farm dwelling. What we would have said was that we were happy with the overall figures we would have published at that time to which the Senator refers but we flagged caution at that time in 2013 about the distribution across the sector.

Mr. Pádraig Dalton

We have changed our sampling approach because there was a specific sampling issue that gave rise to an element of that issue. Surveys can be subject to sampling errors, and the sampling error for that sector was, unfortunately, larger than we would normally have. We undertook a methodological project. Our statistical methods and development team undertook a review of the quarterly national household survey, QNHS sample and we have now changed the sampling methodology behind the QNHS to ensure that those types of issues will not arise again.

Senator Rose Conway-Walsh

There was a change at the time, in terms of the status of farming within the Department of Agriculture, Food and the Marine that influenced that in some way. It comes back to whether there is a role for the CSO in terms of information, knowledge and recommendations that can be made beyond producing statistics. I ask Mr. Dalton to outline if he sees a role there and what is needed for it to be able to perform a more analytical role.

Mr. Pádraig Dalton

I point the Senator towards the National Statistics Board strategy, which provides strategic guidance to the CSO. The role of the CSO changed in 2015 via two mechanisms. One was through the change in EU law, which means the director general of the CSO is no longer responsible solely for statistics compiled by the CSO. I am now responsible for European statistics compiled right across the public sector. There are 15 other agencies in the State that currently compile European statistics. If there is an issue with the European statistics now compiled by one of those entities, EUROSTAT will come to me and hold me responsible for addressing the issue.

The National Statistics Board strategy has also stated the CSO should be co-ordinating across the public sector to put in place structures and strategies to ensure that data collected for a primary purpose cannot subsequently be used for secondary processing in support of statistical products. We have put a new directorate in place, our statistical co-ordination directorate, and are actively engaging across the broader public service to develop skill sets and data infrastructures that will allow all of us to make better use of the information that has already been collected and that citizens have already provided to us in order that we can provide better evidence to inform decision-making right across the system. I am sure it will take time, but it is a process that is well under way. We have had a lot of very positive engagement right across the system.

Senator Rose Conway-Walsh

I very much welcome that. This committee has done a lot of work on insurance and on motor insurance in particular. We suggested some time ago that the CSO might take on a more proactive role in the collation and publication of data related to insurance premiums and awards. Is the CSO examining what role it could play in this area?

Mr. Pádraig Dalton

This came up in other fora recently. There have been suggestions regarding our involvement. As the committee is aware, an expert group on the cost of insurance has been established by the Minister of State, Deputy Eoghan Murphy, and the CSO is sitting on that sub-group. We are actively involved in that sub-group which, I understand, will report by the end of October. We will not pre-empt the findings of that report, but subject to them we will then determine how best the CSO can support the provision of more evidence in this area.

It is important to remember the mandate and role of the CFO. When we become involved, the information we compile legally can only be used for statistical purposes. It cannot be used for regulatory or administrative purposes. Of course, we can only put aggregate statistical data into the public domain. The CSO can provide support, but only within the framework of our mandate. We are happy to provide support and can see that we would have a role to play, but the main question is on what is the appropriate data source for some of this information, where is it best gathered and how is it subsequently best exploited to inform discussions. We can give our expertise and input into that discussion very constructively.

Senator Rose Conway-Walsh

Okay. I thank Mr. Dalton.

Senator Paddy Burke

I welcome the delegation. The presentation was very interesting. Businesses have no choice but to provide information. Is information sought from the same businesses year after year?

Mr. Pádraig Dalton

It varies. Obviously, large multinational enterprises are required to provide information and they would be included in all samples. Small and medium-sized enterprises are not included in samples every year. We are continuously moving towards the broader exploitation of administrative data to support the production of our statistical data. For example, in our annual services inquiry, we no longer survey enterprises with fewer than ten persons engaged. We get the data from an administrative data source and use modelling techniques to produce information for that particular sector. We produce a response burden barometer every year. We have reduced the burden on small and medium-sized enterprises over the past number of years, and will continue to make the shift from primary data collection for small enterprises towards the exploitation of existing data sources, in particular administrative data sources.

Senator Paddy Burke

Can individuals and householders opt out of giving information if they so wish?

Mr. Pádraig Dalton

The provision of data to Europe is mandatory. It is a legal requirement on Ireland Inc. The CSO fulfils the role of providing information to Europe. The decision of householders to engage with the survey is voluntary. That can present a particular challenge. If response rates continue to fall, we will have difficulty in providing data. The only place where there is a requirement on individuals to comply with our request is the census of population. It is a legal requirement on every individual in the State. That is the only case where an individual person is legally required to engage with our household surveys and personnel.

Senator Paddy Burke

Mr. Dalton said in some cases it could take more than an hour to complete a survey. Should there be a payment for contributing?

Mr. Pádraig Dalton

It varies from survey to survey. Generally, the CSO does not pay anyone for providing information to fulfil our legal obligations. For example, the survey on living conditions can take up to an hour to complete. The QNHS can take up to 20 or 25 minutes to complete. The most difficult survey we currently carry out is the household budget survey, and we make a small contribution to households because they keep a diary for us for about two weeks and record information on every item of expenditure which, of course, we subsequently use to weight the CPI, which is critical for all of us.

Senator Paddy Burke

What kind of contribution? Is it monetary?

Mr. Pádraig Dalton

No, it is normally in the form of a voucher.

Senator Paddy Burke

Okay. If it was advertised, more people might-----

Mr. Pádraig Dalton

It is definitely advertised when we are trying to recruit householders but householders still must make a significant contribution in terms of their time.

Senator Paddy Burke

What happens in regard to false information? Is the CSO conscious that some information could be false? Can it prove whether information is false or accurate?

Mr. Pádraig Dalton

One of the ways we check for false information is looking at trends across time. If there are any unusual patterns, that obviously raises flags. We have access to a broad range of data sources. For every statistic we compile, we examine a broad range of indicators and use that to ensure the information we are getting is broadly consistent with what one would expect.

Senator Paddy Burke

Is all of the information broken down into regions, such as county by county, or on a population basis?

Mr. Pádraig Dalton

It is our standard practice to produce regional breakdowns of data. In some cases, it is not always possible to produce regional breakdowns, in particular where statistics are based on a sample survey. I do not want to get too technical, but sometimes at county level the sample size is too small to be able to produce reliable and consistent data over time. One could see a change, period-on-period, for example, for any particular county. The change may not be an underlying change.

It could be down to the fact that the sample is too small. Therefore, whereas to the greatest extent possible we will always produce regional data, in some cases we will not be able to do so. The best source is the census of population, through which we produce the small area population statistics. I do not know whether any members have had the time to look at the preliminary census results we published earlier this year. We were able to produce details on vacant houses and vacancy rates down to electoral division level, so it is possible to go into an interactive map, click on any electoral division and be given the number of vacant houses.

A key part of our strategy is to try to build information from the bottom up, that is, collected at the most granular level of detail which, for us, is the Eircode postcode. The Eircode postcode is absolutely essential to our ability to compile and provide information and evidence to policymakers or decision makers at the smallest level of detail possible or appropriate. Vacancy rates are a prime example. It is very useful for policymakers to be able not just to talk about vacancy rates at county level, but also down to small area level. This supports potentially targeted policy interventions, as distinct from broad brushstroke interventions, into the future. Targeted policy interventions are obviously more cost-effective, and outcomes can be more effective-----

Senator Paddy Burke

How does the CSO get that information? Does it get it through the census?

Mr. Pádraig Dalton

Again, everywhere we collect information now, we ask people to provide us with their Eircode postcodes. For the census we were able to get the Eircode postcode at the point of collection so we will now be able to produce the broad range of statistics we collected from the census down to a variety of regional breakdowns, from electoral division area up to nomenclature of territorial units for statistics, NUTS 3 level, namely, county level, and NUTS 2 level. We provide data by, for example, diocese and Garda region, so if somebody provides us with the digitised boundaries, we can build any regional structure from the bottom up because we have the Eircode postcode. We now do this across all our surveys. The Eircode postcode is now part of the core data collection within the CSO and provides us with the potential to produce far more regional data than we were able to produce previously.

Senator Gerry Horkan

I join most of the previous speakers in thanking Mr. Dalton for all the work the CSO does. I spent 12 and a half years on a local authority and we were always waiting for movements in population and information as to where we would either gain or lose territory, which is all dependent on population.

Chairman

And votes.

Senator Gerry Horkan

Exactly. One means the other. The morning the CSO's statistics on the population came out, it was the first topic raised in the Seanad because quite a number of Senators were aspiring to reclaim seats or considering territory where they might have a better chance the next time than the last time because of the CSO's figures.

Regarding compliance rates and response rates, I do not doubt that the 75 large cases are set up in such a way that they engage with the CSO all the time. However, what happens in that regard? I cannot imagine that the CSO gets a 100% response rate from household surveys and so on. Will Mr. Dalton give an indication as to what the compliance rates and the response rates are? Does the CSO do any auditing? If somebody fills in a form and writes something, how does the CSO know whether it is accurate? Does the CSO trust the person or does it have metrics and look for outliers? How does it all work?

Mr. Pádraig Dalton

I will take the last question about how we know whether the data are accurate, first. Particularly on the business statistics side, we have access to a very broad range of information. Regarding the 75 multinational enterprises covered in the large cases unit, we look right across a broad range of data sources. Mr. Connolly and his team pull information from the balance of payments survey; all the annual business statistical surveys we conduct in Cork; the monthly and quarterly statistics, which we call the short-term statistics; the trade data, to which we have access; and the corporation tax data, to which we also have a legal right of access. Therefore, for any given entity, we will look across the range of information and combine the data at source, and they will either paint a consistent picture or they will not. If they do not, we must try to work out whether the discrepancy is explainable. If it is not, we visit the company and tell its representatives that something is not adding up and ask them to explain to us the situation.

That is what happens in the case of the large multinational companies and that is how we can be absolutely sure that we have very high-quality data for them. For many of the other businesses, we are able to cross-check on some variables. The administrative data sources, such as corporation tax, do not always contain all the variables we need, which is part of the problem, but they do contain some variables that we can use to quality-check what the businesses themselves have provided to us. That quality check across different sources is very important for us. In many countries the balance of payments is not compiled by the national statistical institute, but by the central bank. I think there are only four or five countries in Europe in which the national statistical institute compiles the BOP. This is a real advantage for us, particularly when we are trying to compile things like national accounts, because we need access to as much data as possible. That is how we verify the data.

The first question concerned compliance rates. Compliance rates are quite good right across the board. Regarding the compliance rate in our survey of households, we have a very good, professional field force. We have about 100 staff, our permanent field force, who are out knocking on doors every day of the week, 52 weeks of the year. It is a very professional group of people that does an extremely difficult job for us very well. They have gained much experience over the years of how to gain compliance at the doorstep so, in fairness to them, they get a very low refusal rate. However, it is getting more difficult because people's lives are becoming more difficult, but compliance rates are generally quite good. We prosecute serial offenders on the business side because, as I said, on the business side it is a legal obligation to compile the information. We will take them through the courts because it is important that we get compliance. Our job is to compile accurate, independent statistics for the State. Compliance rates are absolutely essential for that, and the trust that respondents have in us is important. Non-compliance happens very rarely. The compliance rate on the business side is very high as well. We have a record of protecting the identity and confidentiality of the information provided to us. That is a core value of the CSO.

Senator Gerry Horkan

I take Mr. Dalton's point about confidentiality and not being allowed to divulge information that would identify people, but is it fair to say that the 75 companies are all companies of foreign origin, or are indigenous Irish companies, as we would understand them, among the 75?

Mr. Pádraig Dalton

The vast majority of the 75 would be foreign-owned multinational enterprises.

Senator Gerry Horkan

Senator Conway-Walsh referred to motor insurance. Perhaps Mr. Dalton has already answered in speaking about the working group and so on. The entire thrust of our discussions over six or seven days about motor insurance was the lack of available data. Every witness who appeared before the committee said that he or she was giving enough data or that the data were there. Would it be within the CSO's remit, if asked, to compile information on things like motor insurance claims? I am sure it has the capacity to do so. Of the claims, 10% are going to court, 20% are going to the Injuries Board and 70% are being settled. We do not have data on the invisible 70% and are relying instead on people telling us stories. We would like to see data on the number of claims, the increase in claims, the amount paid out per claim, the amount paid in legal fees, the number of claims relating to whiplash and so on. Could that be done by the CSO or does Mr. Dalton think the Central Bank should or could be doing it?

Mr. Pádraig Dalton

One of the prerequisites of our involvement is a data source that we can use to compile the relevant information. At the moment one of the problems seems to be the lack of such a database. In certain sectors we are able to go to an administrative data source in whatever Department and start to compile statistics from it. We are trying to consider the matter in the following way. We are involved in the working group on data needs and, as I said, that group will meet at the end of October. We will wait and see what comes out of that report at the end of October and then we will engage constructively with the outcome of it and work with those involved to identify the most appropriate way to provide information on this very important sector. We are very conscious of the need for this type of information but, again, we can only provide it within the broad scope of our mandate.

The most important thing is we would never get involved, and nor legally can we, in any regulatory or administrative type of activity. If there is a regulatory activity which collects the appropriate data, we can use it as a data source to compile official statistics on the sector. This may be something that comes out of the group. I do not know. We will await recommendations and engage constructively then.

Senator Gerry Horkan

We have had quite a lot of discussion about GNP, GDP and net national income. When I studied first year economics there were differences between GNP, GDP and foreign multinationals. It has got more complex. Mr. Dalton mentioned the shift in GDP was 26% in 2015. It is reasonable to say that if the debt-to-GDP ratio is being spoken about and lauded as falling, it is also falling because GDP is rising. It is a percentage. The debt may not change, but GDP is changing significantly. Stating the debt-to-GDP ratio has improved means it has improved relative to GDP, but this is because GDP has risen by 26%.

Mr. Pádraig Dalton

It is a mathematical equation.

Senator Gerry Horkan

Exactly, so it is fair to say it is the case. With regard to the concerns about EU scrutiny, is there anything the CSO wants us to do as legislators or committee members? The CSO has expressed its concerns.

Mr. Pádraig Dalton

We have no concerns about EU scrutiny. In fact, EU scrutiny helps us-----

Senator Gerry Horkan

I refer to how data are collected and the new EU measures and various new surveys.

Mr. Pádraig Dalton

If we are speaking about the broader conversation, the message to take from here today is we need to change the narrative and broaden the discussion beyond GDP and GNP. This is not something that can be achieved solely in the statistical world. It is very important this conversation is broadened to the policy side and the legislator side, particularly at European level, to support the Commission, the IMF and the UN in their work in trying to broaden the discussion on globalisation. This is perhaps where the most support can be given.

Senator Gerry Horkan

It is fair to say that in a small open economy like Ireland's, where so much of our activity is internationally traded, be it goods or services, GNP and GDP effectively are not fit for purpose in terms of being the figures with which we should assess our performance.

Mr. Pádraig Dalton

I want to be a little bit nuanced. What I would say is GDP and GNP do not provide useful insights into what is happening in the domestic economy. GDP and GNP are the international standards on the performance of the economy as a whole. As somebody described it one day, GDP and GNP accurately reflect the Irish economy but they do not give much insight into what is happening in the Irish economy. This is a slight nuance but it is quite important. It is important that we all recognise that as things stand, GDP and GNP are just two of the international standards, but the CSO must continue to compile GDP and GNP in accordance with the standards. We absolutely recognise the need to move beyond GDP and GNP. We are working in this direction and the expert group will help us in this realm.

Chairman

I want to go back to a point raised by Senators Horkan and Conway-Walsh on the insurance industry. Insurance Ireland tells us it holds a significant amount of information on the industry, which is available. Is it within the remit of the CSO to conduct a survey on issues relative to motor insurance, including the reasons for the hike and the type of claims being made, including the 70% on which we do not have visibility? As the CSO would do with some sectors in the economy, would it write to all of them asking for specific information and bring it back in?

Mr. Pádraig Dalton

For that to happen we would have to look at the Statistics Act and check whether it is within our mandate and remit. If it is, then we would need a statutory order to make response to the survey a requirement on the respondents. In this context, there is a mechanism through which it can be done.

Chairman

What gives the CSO the right to take the transport sector, write to everyone involved in transport and ask them to fill in the information?

Mr. Pádraig Dalton

The right comes through EU legislation that requires Ireland Inc. to provide transport data to Europe and the corresponding statutory order which makes reporting and compliance with our survey that collects that information mandatory. The statutory order is based on the Statistics Act 1993.

Chairman

The CSO feels it might not have that order regarding the insurance industry.

Mr. Pádraig Dalton

At present we do not have such an order in place.

Chairman

If that order were in place, it would enable the CSO to engage with brokers, insurance companies and insurance houses. It would be able to do all of this and collect the data that would give us the information on the claims settled by going to court, without going to court and so on.

Mr. Pádraig Dalton

The broader question really is where is the best place to collect the data is. What we really want to avoid is a scenario where the CSO collects data while a regulator collects the exact same data. This would not be efficient or in the taxpayers' interest. The group is looking at exactly these issues as to where is the appropriate place and what is the appropriate mechanism and appropriate data source. Once the group reports back to us, the CSO will actively engage and we will certainly not be found wanting in playing our role in trying to provide information.

Chairman

The regulator in this case is the Central Bank, and it does not have the information. We have not been able to extract the information from all of the players who came before the committee to discuss insurance. The hard data that are required to influence policy on this or to be able to put information out into the public domain for those who might want to enter the market to analyse, are not available. Someone has them but nobody seems to want to ask for them. This is the issue. The CSO would not be duplicating or getting into the area of regulation. It certainly would be getting into the area of giving a clear picture and analysis on the information that is available but which nobody seems to be collecting. An interesting answer Mr. Roux gave when he was here was that he gets the information from Insurance Ireland for which he asks. My question is, what other information is there? That is not the question he was asked. He was asked whether he got all of the information. The answer was that he got what he asked for. Somebody needs to ask all of the questions of Insurance Ireland, and anywhere else the information can be got, to compile such a report. This does not need an in-depth conversation or consideration because the information is there. Really, what Mr. Dalton is saying is that apart from this, it requires the CSO to be given the remit to do it.

Mr. Pádraig Dalton

This is definitely an important conversation. From our perspective, the key questions are on who is the appropriate authority to collect the information and for what will the information be used. If the CSO collects information from anyone, that immediately rules out the ability to use the information for any purpose other than statistical purposes. When we are involved it is quite a limited remit and the outputs are quite limited. All we can do is produce aggregate data. I do not know the answer to this because it is moving into policy and that is uncomfortable because the CSO does not speak on policy. We can speak on data-related issues. The question is whether there is a need for a broader range of data to be used beyond purely statistical purposes. If there is, the question is whether it should be collected elsewhere and then the CSO, through its powers under the Statistics Act, can go to the data source and extract the information it needs to compile aggregate statistical data to inform the debate.

Chairman

That is the point I am making. We are told there are databases in the control of Insurance Ireland that would be helpful to the ministerial group of which the Central Statistics Office is part and this committee.

The Central Statistics Office, as a reputable organisation, is the appropriate body to carry out such an analysis which should be general. The questionnaires sent to businesses are very detailed. I noted the other day while reading one of them that if a similar set of questions was asked of the insurance industry, it would be possible to collect a large amount of usable data.

Mr. Pádraig Dalton

We fully accept that the Central Statistics Office can play a role in producing statistical data to inform decision makers on this important issue. The question is how to do this and whether we should go down the route of conducting an expensive survey to collect the information directly from the relevant entities, which would only be used for statistical purposes, or whether we should take a more holistic approach and develop a more comprehensive data source that would go beyond the pure statistical requirements and meet the requirements of regulation and administration. The second option would allow the CSO to access a centralised database of information collected for a multitude of purposes beyond statistical purposes. We could then compile the aggregate statistical analysis from this data source. That is the only issue and the CSO is awaiting the outcome of the work of the group.

Chairman

Ultimately, the CSO cannot do it because it is not within its remit to do so. Is that the position?

Mr. Pádraig Dalton

We do not have a statutory order in place to do it.

Chairman

That answers one question. The CSO has indicated that it is trying to eliminate the burden of red tape its questionnaires impose on the small and medium enterprise sector. This means that companies with fewer than ten employees do not receive a questionnaire. I have seen the questionnaires sent to companies and assisted some companies in completing them. The way in which questions are asked is not appropriate to some companies, although I presume they are generally appropriate. What engagement does the CSO have with the various sectors to which it sends questionnaires to establish the breakdown within each sector in order that it can make its questionnaires more relevant? For example, in the transport sector a question will relate to whether a company is shifting goods by truck, trailer, van or whatever else and likewise in respect of the retail trade. What effort is being made to ensure questions are framed in such a way that the CSO will obtain the information it wants and avoid a scenario in which someone completes a form because he or she knows that if the company is to be in compliance, the boxes must be ticked and the form returned? This frequently happens and it skews outcomes for the CSO.

Mr. Pádraig Dalton

In case anybody believes the CSO is making up the questions in its office, all of the requirements, particularly in the transport areas to which the Chairman referred, are set down in EU legislation. The European Union prescribes what the CSO and every member state must provide. We must collect the information from the entities.

Chairman

In that case, it is a lost cause.

Mr. Pádraig Dalton

No, it is not a lost cause because we also have a transport statistics liaison group through which we engage with a broad range of users. We also engage with stakeholders to try to work out what is the best way to collect information from respondents. The Central Statistics Office is trying to move away from exclusive reliance on paper based data collection towards a web based data collection approach similar to the Revenue online service for collecting data from businesses. This would be much more efficient and effective for businesses. While it would require up-front investment, it would, over time, further reduce the burden on businesses. We are very conscious of the burden we are placing on businesses and the need to ask questions in a manner that is consistent with the way in which the entities or companies compile their accounts.

Chairman

In the context of Mr. Dalton's point about the European Union, the issue is that one size does not fit all. Now that I know that a questionnaire is compiled by someone in Europe, I can at least understand the reason the questions are difficult to answer for many small and medium enterprises. The difficulty is twofold because, first, completing the questionnaires imposes an unfair burden on small businesses and, second, it is a legal requirement to return completed questionnaires that are specific and accurate. Many companies tick boxes to comply with the legal requirements. The European Union authorities and the Central Statistics Office must be conscious of this and make every effort to ensure businesses are able to answer the questions asked and that they are appropriate to the business in question.

Mr. Pádraig Dalton

I fully appreciate that and we are working to improve the position. Mr. McMahon has worked in this area and may have something to add.

Mr. Richard McMahon

To elaborate on the road freight survey to which the Chairman referred, I certainly acknowledge that it is quite burdensome. The Central Statistics Office does its best to exclude smaller businesses from this inquiry. We set a threshold for the size of goods vehicles, which means that we exclude all good vehicles of less than 2 tonnes unladen weight. In our sampling we are very conscious of the burden imposed on small businesses. We were active participants in the high level group on the regulatory burden and since 2005 have reduced the response burden for small businesses by between 50% and 60%, which is a significant amount. We are particularly conscious of small businesses.

Chairman

I was that soldier, as it were, and having filled in CSO questionnaires, I can tell the delegates that my responses were not always accurate, even if I did the best I could.

Deputy Pearse Doherty

I looked up the quarter two figures for average incomes on the CSO's website yesterday. Will the delegates explain the reason average incomes have been declining across most heads in quarter two compared with quarter one? This trend fits with a fall in revenue under some income tax heads. Is the CSO picking up any of the reasons for the decline in average incomes across different sectors?

Mr. Pádraig Dalton

Perhaps Mr. Connolly might have data for quarter two.

Mr. Michael Connolly

Is the Deputy referring to household income?

Deputy Pearse Doherty

No, I am referring to average wages across various sectors, ranging from industry to the arts, for which quarter two data have been published.

Mr. Richard McMahon

They may include the census of population enumerators who were employed during that period. I do not have the figures to hand, but that would certainly impact on the figures for the first two quarters of the year. We recruited approximately 5,000 temporary staff to conduct the field operation.

Deputy Pearse Doherty

I am not referring to incomes among staff in the Central Statistics Office but average incomes in the economy. The figures are broken down under various headings, including industry, the public sector, banking and financial services, the arts, the community sector and so forth. The figures show a decline in average incomes in most sectors. While more employment is being created, I wonder if the decline in incomes is likely to become a trend.

Mr. Brian Ring

I will revert to the Deputy on the issue. I believe, however, that he is referring to outputs from the earnings, hours and employment costs, EHEC, survey. Specifically, it shows an average weekly earnings figure in quarter one of 2016 of €713 and in quarter two of €703. Perhaps these are the figures to which the Deputy is referring. While there has been a positive annual change of 0.5% in the headline figures, the quarterly change was -0.13%.

The Deputy referred to new employment. There can be compositional impacts on some of the figures because the EHEC survey is broad. The overall average figure can be distorted by compositional impacts.

The CSO, like many public sector bodies, has begun to recruit people again as the moratorium has been lifted. In terms of new rules this is a public sector example but it can relate to many other industries. People who started in recent years were at the bottom of pay scales, etc., and, therefore, there can be compositional impacts on the overall average figures.

The figures for the second quarter of 2015 and the first and second quarters of 2016 are not seasonally adjusted and, therefore, there can also be seasonal impacts on the broad figures. Average weekly earnings are impacted by the number of hours worked. Earnings have been relatively sticky over recent periods. The biggest thing I would be aware of in that regard is an impact of compositional changes.

Deputy Pearse Doherty

Compositional changes should not be seen from quarter to quarter, as opposed to from year to year because there has not been a dramatic increase in employment with people coming in on the first metric in terms of the pay scale. The average industrial wage reduced by just over €30 quarter on quarter, which I imagine is not the result of a large number of people being employed over that three-month period. The trend up until now was a gradual increase in the average wage and there has been an increase in employment over a sustained period, thankfully. There seems to be a correlation between what the CSO is identifying as a reduction in average incomes and the tax receipts. Is there something going in the economy that has not been subject of much debate such as wages being reduced? I am not sure what is happening.

Mr. Pádraig Dalton

We will definitely look into that for the Deputy and try to come back with a comprehensive answer. We always urge people not to just look at monthly figures but to look at the longer-term trend because specific issues can impact on monthly figures. When I was in Mr. Ring's shoes and was responsible for the quarterly national household survey, QNHS, I always used to say that one month does not make a trend. We would always look over the next number of quarters to see if this was the emergence of a trend.

Deputy Pearse Doherty

When will the third quarter figures be released?

Mr. Brian Ring

Approximately 25 November. The next QNHS relating to employment will be released Tuesday, 22 November. We have brought back to publishing at Tplus 51 days to help ourselves and policy makers and consequently we publish on the eighth week. The EHECS generally is released within Tplus 56 days, which is relative to the end of the calendar quarter we are looking at. It is end of November for both of those releases. The Deputy is correct that there is stuff going on. We are aware that the labour market and the economy are dynamic at the moment in terms of change. I agree with Mr. Dalton that the broader evolution of trends across the surveys is something to be conscious of as well.

Chairman

I have three questions. First, I hope the new regulations will not mean the CSO will just become a collection agent for the Commission but will still retain its independent position domestically.

Mr. Pádraig Dalton

Absolutely, yes.

Chairman

Second, the organisation places a great emphasis on the census. Will that be affected in any way?

Mr. Pádraig Dalton

No, that will not be affected but Census 2021 will be a regulatory census. Most countries carry out a census every ten years and, therefore, the next census is a requirement under EU law.

Chairman

But that will not affect the CSO's use of the data.

Mr. Pádraig Dalton

Absolutely not.

Chairman

Third, when the CSO collects all these data, can they be sold on or can they be made available to, say, a business?

Mr. Pádraig Dalton

We spend taxpayers' money and as far we are concerned, official statistics are a public good. Everything we put out, we put out into the public domain. We not only put the data on our website using the e-releases but all data are made available in open linked data format and we are, therefore, compliant with the open data agenda.

Chairman

Anyone can pick it up.

Mr. Pádraig Dalton

Absolutely, yes.

Chairman

There are no further questions. I thank the officials very much. The committee has to decide on follow-up action on these proposals. Is it agreed to engage with the EU Commission to raise the CSO's concerns regarding the proposals, in particular, the methodology proposed to collect the data? Agreed.

The joint committee adjourned at 12.35 p.m. until 10 a.m. on Thursday, 20 October 2016.