Finland

In Finland three surveys on the number of pigs are carried out per year, namely in May, June and December. The surveys carried out in May (reference date 1. 5.) and in December (reference date 1. 12.) are integrated livestock surveys (excluding bovine animals).

The surveys carried out in June and December are sample surveys, while surveys carried out in May are full surveys. The average sampling error is between 1,79 and 2,05 %. The Rural Business Register includes the number of pigs annually.

In 2000, the sampling frame included a total of 83.000 farms, with at least one hectare of cultivated arable land or at least one animal unit. To facilitate sampling the sampling frame was stratified into 407 strata (307 strata plus 100 biggest farms, each one for own strata). Stratification variables were farm's geographical location (16 Development Centers), size of the farm (calculatory unit that was a mean of animal unit and crop area on the farm, divided to 4 categories) and farm typology (7 categories). The size of the farm was the allocation variable. Allocation was carried out by using the Neyman allocation. All those farms included in the stratum with less than 10 farms are within the sample. The sample size was about 10.000 farms. The sample is a panel sample of which one-third of the farms is drawn afresh each year, based on the aforementioned allocation method.

By the sample surveys 46 % of pig population is recorded. The survey covers 2.000 farms which is 44 % of all farms breeding pigs.

The three surveys are carried out in selected regions. The survey is carried out by Employment and Economic Development Centres in all regions.

The method used for estimating the pig population which is not recorded by the sample surveys in June and December is the following: Estimates are calculated on the basis of sample using the CLAN-program developed by Statistics Sweden. The estimation method used in the survey is Horwitz-Thompson type of estimation. Data is chiefly calculated by using the normal estimation formula for stratified sampling. For comparison, data is also calculated by means of combined ratio estimation. Woodroff's transformation is used for calculating standard errors. Comparing standard errors show which method of estimation produces more accurate data.

In June and December data are obtained by oral interviews with farmers and by telephone. Data for surveys in May are obtained from Rural Business Register. The average response rate is 80 % (in June and December surveys). The Information Centre of the Ministry of Agriculture and Forestry is responsible for carrying out the survey.

Finland plans to make use of administrative sources in the medium term (one to five years). The data source available for this is the Rural Business Register.

Surveys on the number of cattle

In Finland administrative data are used to get the livestock for cattle. For this purpose one has the Bovine Animal Register, which is a part of the Rural Business Register. The data are extracted twice per year, namely in May and December.

The Ministry of Agriculture and Forestry and the Agriculture Accounting Centre are responsible for extracting the data.

Surveys on the number of sheep and goats

In Finland two surveys on the number of sheep and goats are carried out per year, namely in May and December. Sheep and goat populations are recorded separately. All the surveys are integrated livestock surveys.

The surveys carried out in December are sample surveys, while surveys carried out in May are full surveys. The average sampling error in December surveys is between 5,2 and 10,0 % for sheep and between 9,5 and 18,2 % for goats. The last full survey was carried out in May 2000.

In 2000, the sampling frame included a total of 83.000 farms, with at least one hectare of cultivated arable land or at least one animal unit. To facilitate sampling the sampling frame was stratified into 407 strata (307 strata plus 100 biggest farms, each one for own strata). Stratification variables were farm's geographical location (16 Development Centers), size of the farm (calculatory unit that was a mean of animal unit and crop area on the farm, divided to 4 categories) and farm typology (7 categories). The size of the farm was the allocation variable. Allocation is carried out by using the Neyman allocation. All those farms included in the stratum with less than 10 farms are within the sample. The sample size was about 10.000 farms. The sample is a panel sample of which one-third of the farms is drawn afresh each year, based on the aforementioned allocation method.

By the sample surveys 15 % of sheep population and 16 % of goat population are recorded. The survey covers 259 farms (12 %) breeding sheep and 109 farms (16 %) breeding goats.

The two surveys are carried out in selected regions. The survey is carried out by Employment and Economic Development Centres in all regions.

The method used for estimating the sheep and goat population which is not recorded by the sample surveys in December is the following: Estimates are calculated on the basis of sample using the CLAN-program developed by Statistics Sweden. The estimation method used in the survey is Horwitz-Thompson type of estimation. Data is chiefly calculated by using the normal estimation formula for stratified sampling. For comparison, data is also calculated by means of combined ratio estimation. Woodroff's transformation is used for calculating standard errors. Comparing standard errors show which method of estimation produces more accurate data.

In December data are obtained by oral interviews with farmers and by telephone. Data for surveys in May are obtained from Rural Business Register. The average response rate is 80 % (in December surveys). The Information Centre of the Ministry of Agriculture and Forestry is responsible for carrying out the survey.

Finland plans to make use of administrative sources in the medium term (one to five years). The data source available for this is the Rural Business Register.

Notes on the livestock surveys

In Finland, two sample surveys on number of livestock are carried out every year, as of 1st of June and as of 1st of December. These surveys are integrated livestock and crop surveys. The Rural Business Register includes data on all number of livestock as of 1st of May. The Bovine Animal Register is part of the Rural Business Register. Number of cattle is not obtained by the surveys, but data is taken from the Bovine Animal Register twice a year.

Statistics on slaughterings

Monthly slaughtering data are available in Finland on the number and slaughterweight for pigs overall, calves, heifers, cows, bulls, bullocks, sheep overall, lambs and goats overall.


	Number	Slaughter Weight
Pigs, total	X	X
Calves	X	X
Heifers	X	X
Cows	X	X
Bulls	X	X
Bullocks	X	X
Sheep, total	X	X
Lambs	X	X
Goats, total	X	X

Slaughtering data is recorded monthly by the Information Centre of the Ministry of Agriculture and Forestry. Basic rule is that slaughterhouses or other kind of enterprises which has bought animals from farmers report it to the Information Centre of the Ministry of Agriculture and Forestry. All the enterprises report the number and the weight of animals slaughtered in abattoirs. Big enterprises report also the prices paid to farmers. Monthly slaughtering data is available within two months after the end of the reference month. Data on slaughtering on farms are based on the sample survey carried out by the Information Centre in June and December.

Data on external trade in live animals

Foreign trade of live animals is recorded by customs statistics. Trader report the volumes traded to customs officers. However the amount of live animals traded is very small and it does not affect GIP-numbers at all.

Production forecasts (gross indigenous production)

number of animals from the surveys of TIKE, the Rural Business Register

monthly slaughter statistics by type of animals (kg and no.) from TIKE

carcass statistics by municipality from TIKE

milk production models of Gallup ETT

the purchasing network of the slaughtering companies

the producer surveys of Gallup ETT

The number of animals on the farms (TIKE surveys, Rural Business Register and the producer surveys of Gallup Food and Farm Facts (Gallup ETT)) are compared with and connected to the slaughter statistics of TIKE.

For the beef meat forecast the following data is used: the number of lactating cows, the normal monthly slaughtering, the number of calves born, and the number of inseminations of cows. These data are used for forecasting both young beef and cow meat coming to the markets. The data are collected from the milk-production forecasting models of Gallup ETT and the slaughter statistics of TIKE. Also the Rural Business Register and the producer surveys of Gallup ETT.

The pig meat forecast is made on the basis of data from the sample surveys of TIKE and the Rural Business Register. The meat coming to the markets in the coming months is forecasted using slaughter data from TIKE, the number of piglets distributed through the companies, and the relation between the distributed piglets and slaughtered animals (part of combined production). The developing trends are also recorded through the producer surveys of Gallup ETT.

Poultry statistics

Slaughtering data are collected also in the area of poultry statistics. Monthly slaughtering data on broilers and turkeys and also monthly slaughtering data on hens, ducks, geese and other poultry are available. Amount of ducks, geese and other poultry is so small that they are reported all together. Enterprises are reporting on carcasses so coefficients are not used.

It might be possible to have an econometric model to provide information on the poultry population. At this moment there is not such a model available. Data on poultry population is available from the Rural Business Register (reference data 1st of May).

By the sample surveys carried out by the Information Centre in June and December information on the number of laying hens and the production of eggs is obtained as well.