This week farmers in the UK and Ireland are labouring to complete an agricultural census for the European Union. The questionnaire can be filled in online or as a paper copy, and completion is mandatory. (All member states are required to complete it, but timing varies from state to state.)

The questions are designed to provide a comprehensive picture of agriculture across the EU: land areas, crops, livestock numbers, employment, even how manure and slurry are stored on farms and how pigs and poultry are housed. The census is carried out every ten years, and in England will cover 120,000 holdings – farms smaller than 5 hectares of agricultural land or goatherds having charge of fewer than 20 goats (to take two examples at random) are excluded.

The EU census is in addition to the annual June and December Agricultural Censuses, which despite their names are large national surveys covering some 40,000 holdings and asking 150 questions. Farmers also contribute data to a bewildering range of crop and livestock statistics, many for the EU: UK Wheat Milled and Flour production (monthly), Cereals Usage by Oatmeal Millers in the UK (annual), GB Production of Bacon and Ham (quarterly) and UK Slaughter Statistics – Clean Pigs (quarterly), to give just a flavour.

The question is: are these statistics really necessary? In a recent assessment of DEFRA statistics, the UK Statistics Authority questioned whether the £217,000 annual cost of the crop and livestock statistics was really justified. (No cost was identified for the 2010 census.) DEFRA admits that many of the statistics produced for the EU have a narrow audience in the UK. Nor does the European Commission document the use made of them or the decisions they help to inform. The authority says that without these justifications the statistics cannot be designated as National Statistics.

The chair of the authority, Sir Michael Scholar wrote to the agricultural commissioner, Dacian Ciolos, to try to discover what use, if any, was made of all this data. Agriculture accounts for less than 1 per cent of GDP and 1.7 per cent of employment, he noted, yet a full census with very detailed labour requirements was being held. Changes to the common agricultural policy, with reduced emphasis on market management and production support did not appear to have been reflected in any change in the requirements for statistical information.

Mr Ciolos turned the letter over to Jean-Luc Demarty, head of the Directorate-General for Agricultural and Rural Development. His reply (which together with Sir Michael’s letter is available on the authority’s website) claims that the EC is “firmly devoted to striking the right balance between information needs, legislative requirements and administrative burden for the member states.”

The issue is on the table at a meeting this month of the Simplification Experts’ Group, he adds, and the proposed reform of CAP in 2013 will provide another opportunity for taking stock. Other possible improvements include greater use of administrative sources and re-assessing user needs to “rebalance” the current demands of (sic) agricultural statistics data.

I think that last point means reconsidering whether we need all the data that farmers are pumping out, but I don’t speak Euroish, so I’m not sure. Specialists will already have read the Commission Communication of last August (COM (2009) 404) which explains how the European statistics machine is to be reformed.

It’s not to be recommended for non-specialists (maybe the Simplification Experts' Group could take a look at it) , but it does admit that the present system imposes “an unnecessarily heavy burden on respondents”. Farmers sweating over their questionnaires this week, rather than their harvests, will doubtless agree.