DAFM: Graham Neilan, John Muldowney, Sinead McPhillips, Wila Bruce, Nick Barwise
CSO: Paul J Crowley, Kathryn Foskin, Colm Hassett, Gerry Reilly, Beryl Cronin, Viacheslav Voronovich (CSO-Agriculture); Gerry Brady, Reamonn McKeever, Nova Sharkey (CSO – Environment);Ciaran Counihan (CSO-Trade);Carmel Hinchion , Anthony Dawson (CSO-Prices)
Teagasc: Kevin Hanrahan, Brian Moran
EPA: Phillip O’Brien
Bord Bia: Eoin Kelly
Apologies: Joe Treacy (CSO), Bernard Hyde (EPA), Derry Dillon (Macra), Rowena Dwyer (IFA), Alan Renwick (UCD), Jane Brogan (EPA)
Paul J Crowley (CSO) opened the meeting. The minutes from the last meeting were adopted. Paul outlined the agenda for the meeting and it was adopted. He then gave a brief overview of activities in the CSO since last year, including the publication of FSS 2013, the production of the meat supply balance, the developments in the use of administrative data and the introduction of web survey collection in the pig survey. Work on FSS 2016 was well underway and questionnaires will be issued to 55,000 in the last week of May. Other items of note included methodological changes implemented in Agriculture Accounts and the on-going work on the development of the EU Strategy on Agriculture Statistics, 2020 and beyond.
Paul Crowley gave a presentation on the measurement of employment in agriculture. He explained why the QNHS measurement of employment in agriculture differs from the numbers working in Agriculture according the Farm Structure Survey. He clarified that QNHS is the official measure of employment in Ireland. He has produced a draft information note on employment in agriculture, which will be published on the CSO website, pending comment from ASLG members.
Kevin Hanrahan asked about the issue of trends in the QNHS data as opposed to absolute numbers. The CSO explained the sampling issue in the QNHS in the 2008-2009 period, as had been discussed at the last ASLG, and that this note was aimed at being a note for the present and going forward to explain the difference between QNHS and FSS figures. Kevin expressed the importance of accurate historical data, and asked would there be a possibility of revising the QNHS data time series. The CSO said it would provide DAFM with an age-profile of these 70,000. The CSO said they would bring this request to the QNHS team.
Measuring the viability of farms was discussed, particularly considering the 70,000 farmers where farming is their sole occupation. Questions about the age profile and the consideration of pension income in the calculation of Standard Output were asked. CSO said that this analysis doesn’t account for income as it just focuses on employment data, and that this might be best discussed in terms of the later discussion on profiling farm households. The draft note identifies some key issues which help to explain some of the reasons (farm size, standard output, labour input, and significance of farming) for differences in employment. It agreed to include a statement in the note stating that it was excluding other factors which are more difficult to quantify.
Kathryn Foskin (CSO) presented developments in the Farm Structure Survey and the increasing use of administrative data. FSS 2013 was the first time the administrative data was combined with a sample survey in order to create a census file for most of the required characteristics. It was explained how the usage of administrative data has freed up space in the questionnaire to meet emerging data needs, improve data quality and helped to reduce the burden on farmers. However, the timely provision of administrative data is essential for the process to operate efficiently. The draft FSS 2016 questionnaire was circulated to the group and key parts of the questionnaire (changes) were highlighted in the PowerPoint presentation. Kathryn mentioned that there has been a change to the coordination role of the CSO brought about by Regulation (EU) 2015/759 on European Statistics last year, which formally assigns responsibility to the CSO for coordination of the compilation of European statistics by other national authorities (ONAs). In addition, the NSB Strategy for Statistics 2015-2020 prioritises the development of the Irish Statistical System (ISS) and CSO’s role within it. Members were informed that over the next few years the CSO will be taking a more active role in the development of the ISS and will be directing specific resources to this.
There were a few requests for additional questions in the FSS questionnaire. Flowers are available on the Basic Payment Scheme (BPS) and so will not be collected on the questionnaire. The possible inclusion of an Eircode question was discussed and the CSO said they would consider it. DAFM mentioned that the precise length of long-term leases would be useful. They also asked how a respondent would fill the form in if the farm were jointly managed. The CSO said they would consider this for the information note that will accompany the questionnaire.
Gerry Reilly (CSO) outlined developments relating tothe European Strategy for Agricultural Statistics 2020 and Beyond, which involves the introduction of two new framework regulations, one for Integrated Farm Statistics (IFS), and one for Statistics on Agricultural Input and Output (SAIO), with a focus on coherence across the regulations.IFS will replace the farm structure surveys and include parts of agri-environmental statistics presently not under legislation where data needs to be collected at farm level (e.g. irrigation, manure or nutrient use). All data under IFS will be transmitted to Eurostat as microdata.
SAIO will consist of aggregated crop and animal production statistics, agri-environmental statistics on fertilisers, nutrient balances and pesticides, and potentially agricultural price statistics. All these data are aggregated statistics with no microdata transmission to Eurostat and deal with agricultural inputs (prices of seeds, pesticides, feed etc.) and outputs (crop and animal production and prices).
John Muldowney (DAFM) asked about measuring throughput of cattle in housing units in the Animal Housing module. He also noted that in terms of the irrigation module, water usage in Ireland would be on livestock farms as opposed to the irrigation of land. CSO explained that the characteristics for the modules in the IFS have not been specified yet, but they will consider these factors in the specifications.
Sinead McPhillips (DAFM) outlined the new strategic plan for the development of agri-food sector over the next decade. She presented the long term vision of the plan (“Local Roots, Global Reach”) and highlighted the key growth projections over the period, particularly an 85% increase in exports to €19 billion and the delivery of 23,000 additional jobs to the sector. The implementation of this plan through over 400 action points is detailed in a Food Wise Implementation Plan, which is overseen by a High Level Implementation Committee.
Kevin Hanarahan asked on what basis the 23,000 additional jobs would be measured, again highlighting the importance of an accurate measure of employment in the sector. DAFM mentioned that CIP is used as a basis for one of the indicators, namely GVA. There has been no CIP publication since the 2012 data and DAFM had asked for clarification when the CIP data series would be made available again. Paul Crowley indicated that the Division concerned are currently engaged in finalising an extensive methodological and data sources review for SBS and Business Demography data. There was no firm date at present, but when revised a new SBS and Business Demography series for 2008-14 will be published. It was the intention to finalise this work this summer with publication to follow sometime thereafter.
Paul J Crowley asked if the targets set out in the DAFM plan were considered achievable. Sinead responded that lots of challenges existed but that the targets were achievable and that previous strategies achieved their targets.
The projections made in Foodwise 2025 highlighted the need for clear baseline figures on employment in agriculture and food (QNHS) and farm holders (FSS / Census of Agriculture) and exports (Trade data).
This was a proposed discussion topic from DAFM. They mentioned some potential anomalies (e.g. beef exports to China despite lack of beef access that were occurring in CSO Trade data that they would like to have explained. DAFM receive a monthly breakdown of agriculture imports and exports which comprises approximated 2,000 CN codes divided into 15/16 categories. This information is used by DAFM internally and also as a basis for ministerial briefings, press queries, Parliamentary Questions, etc. Divisions within DAFM who would have knowledge of the various markets have queried some of the trade figures. Differences between CSO and Bord Bia data were also mentioned. DAFM also enquired whether it would be useful for the Department to issue a note to companies stressing the importance accurately completing trade returns,
Ciaran Counihan was in attendance from CSO Trade section and responded to this query. He highlighted that the CSO trade section compiles data on 9 – 10 billion euro worth of exports per month and that this data is based on data which is collected by the VIMA branch of the Revenue Commissioners on behalf of the CSO. Given the volume of exports it is generally not possible for the CSO to spot anomalies involving relatively small amounts, and so the CSO often only become aware of these anomalies when they are highlighted by data users, and so the CSO is grateful to data users for their input.
Ciaran explained how these anomalies occur and how they are investigated. Once an issue is highlighted Trade Section contact Revenue, who in turn contact the relevant companies. He said that most discrepancies are as a result of either coding errors or timing issues. With timing issues, there is often a delay across a month or two in when a particular transaction is reported, but generally the differences tend to balance out over the course of the year.
With reference to the case referred to by the Department, the addresses on the customs forms were in Hong Kong, but the country code used by the trader was China. A second error had occurred whereby the CN code used for the goods related to beef offal, but the description of the goods on the customs form was of pork offal. CSO intends to revise its data to reflect the correct situation.
Eoin Kelly (Bord Bia) volunteered to meet with CSO Trade quarterly to discuss these issues, and Teagasc and DAFM were interested in participating also. He asked if there was a set methodology for production of trade data defined by Eurostat. He also pointed out the UK / EU trade can have greater anomalies as International Trade is required to be certified and this results in it being more accurate.
Kevin Hanrahan (Teagasc) presented on the topic of Land Prices and Rents. He highlighted the importance of land market data in analysing structural change in the agricultural land sector and the difficulties of conducting this type of analysis in the absence of official data. Ireland has the lowest rental share in the EU (approximately 18%), which limits potential for structural change. He outlined the methodology and results of the joint SCSI/Teagasc Land Market Survey and noted the difference between their results and the recently published CSO data. The conclusion was that they need some more detailed information on the methodology employed in the CSO analysis in order to compare the figures, and they also need to get a better understanding of the relationship between the new series and the old CSO/Eurostat series (1974-2005). This DAFM funded research is now in final year and Teagasc will be organising a meeting of stakeholders to discuss research undertaken. The CSO asked about the sample size, whether land sales with residences were included and whether the reason for transfer was considered. In the CSO analysis, only transactions that are sold on for agricultural use are collected, whereas this study did not account for the use of the land after purchase. The annual sample sizes in the Teagasc survey were approximately between 65 and 90. Land with or without residences were included, which differs from the CSO analysis, which only includes land without residences.
Phillip O’Brien (EPA) presented a study of carbon emission estimates based on cropland activity since 2000 using LPIS data. Activity on a specific parcel can be tracked each year to capture movements between cropping and grassland. Cropland includes crops and temporary grassland managed as part of crop rotation systems.
CSO Environment section expressed a desire to meet with the EPA team regarding Habitat and Ecosystem (environmental) accounting and pointed out that the SEAI (Sustainable Energy Authority of Ireland) are a source of data on energy demand.
DAFM requested a discussion on the possibility of building up a profile of farm households from available data sources. Currently there is a lack of data on income for farm households in particular, which is an issue in measuring the viability of farm households and justifying schemes. CSO explained how the Survey of Income and Living Conditions (SILC) could be used to some extent and explained the methodology in place for estimation. All participants in SILC are asked for their PPS number and if they are farmers they are also asked for their Herd Number. P35 and Form 11 returns are then used for analysis, a methodology inherited from the ESRI. However, the overall sample size for farm households is small. FADN (NFS) was mentioned, but the European Court of Auditors has ruled that income is out of scope for FADN. It was suggested that this could be proposed to the Administrative Data Centre (ADC) in the CSO to consider as a possible project. DAFM asked if it would be possible to publish SILC data on Farm Households, and it was highlighted the SILC publish data on Rural and Non-Rural households.
Last year there was a request through the ASLG for the CSO to revive the Meat Supply Balance data series, and this was published in November 2015. This year, the Milk Supply Balance was requested. This is a more complicated process, but the CSO are willing to look into the production of this data. It will require external expertise, and DAFM said their milk experts would like to engage in the process. There is also an EU requirement for Supply Balance on Crops, to take effect from 2018. Therefore, any progress with respect to Milk, was unlikely until next year.
The update on publication of Census of Industrial Production and other Structural Business Statistics was already covered in the discussion on Food Wise 2025.
Paul Crowley thanked everyone for their active participation in a very productive meeting. A number of interesting topics were discussed. Draft minutes would be circulated by the end of the following week for comments.
The following are the action points arising from the meeting: