The Netherlands

In the Netherlands three surveys on the number of pigs are carried out per year, namely in April, August and December. The surveys carried out in August and December are separate pig surveys, while the one in April is carried out as structure survey. Till 1992 there were three sample surveys in January, April and August and the structure survey was executed in May, from 1993 till 1996 there were also three sample surveys but the reference dates were the first of December, April and August and the structure survey had it's reference date in still inMay.

The surveys carried out in August and December are sample surveys, while the structure surveys in April are full surveys. The average sampling error is less than 1%. The last full survey was carried out in April 2001.

The sample is drawn from the agricultural structure survey. It is a stratified sample. The stratification is based on the type of the holding and on the size (in Dutch size units). The results are calculated by a ratio-estimate.

The sample records 100 % of the total pig population. The sample excludes no holding with pigs, so every holding has a chance to be in the sample. (Holdings in the sample have together about 3,5 million pigs, about 2/3 of the forms are usable, that means 2,3 million pigs in the sample represent 13,5 million in the total population). About 3.000 holdings get a form, and about 2.000 holdings are used for making the figures. The number of holdings with pigs is about 15.000. So the percentage of total holdings is about 20%.

Data are obtained by written inquiries. The average response rate is 66 %. It is known that about 2/3 of the sent forms will be usable. So the number of holdings in the sample is 3/2 of the number that is needed. One believes that not responding holdings have not a big influence on the result, because one can compare it with the sample frame that is updated once a year, and one also uses a stratification that will prevent that a difference in respond rate for different type of holdings will influence the results. Statistics Netherlands is responsible for carrying out the survey.

The Netherlands do not plan to use administrative sources instead of statistical surveys in the medium term (one to five years).

Surveys on the number of cattle

In the Netherlands two surveys on the number of cattle are carried out per year, namley in April and November. The surveys carried out in November are separate cattle surveys, while the surveys in April are carried out as structure survey. Till 1992 there were two sample surveys, with reference date in January and July, and the reference date for the structure survey was May. From 1993 till 1998 one sample survey was carried out in January, and also figures from the structure survey in May were available.

The surveys carried out in November are sample surveys, while the structure surveys in April are full surveys. The average sampling error is less than 1%. The last full survey was carried out in April 2001.

The sample records 100 % of the total cattle population. The sample excludes no holding with cattle, so every holding has a chance to be in the sample. (Holdings in the sample have together about 0,3 million cattle, about 2/3 of the forms are usable, that means 0,2 million cattle in the sample represent 4,1 million in the total population). About 3.000 holdings get a form, and about 2.000 holdings are used for making the figures. The number of holdings with cattle is about 46 000. So the percentage of total holdings is about 7%.

The Netherlands plan eventually to use administrative sources instead of statistical surveys in the medium term (one to five years). It depends on the results of the investigations. The data source available for this is the System for Identification and Registration of Bovine Animals.

Surveys on the number of sheep and goats

In the Netherlands one survey on the number of sheep and goats is carried out per year, namely in April. Sheep and goat populations are recorded separately. The survey is carried out within the structure survey in April. In the structure survey lambs, ewes and rams and milkgoats and other goats are distinguished. Till 1996 the structure survey was carried out in May.

Statistics on slaughterings

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


	Number	Slaughter Weight
Pigs, total	X	X
Calves	X	X
Heifers	X	X
Cows	X	X
Bulls1	X	X
Bullocks2	X	X
Sheep, total	X	X
Lambs	X	X
Goats, total	X	X

1 together with bullocks

2 together with bulls

The public inspection service for livestock and meat (RVV) collects the data on the number of slaughtering of every slaughterhouse and butcher who are allowed to slaughter by their owns. RVV sends these figures to Statistics Netherlands. The Product board for Livestock, Meat and Eggs records the average weight of the animals, and sends also the results to Statistics Netherlands. Statistics Netherlands integrates these figures. The data are available two months after the reference period.

Data on external trade in live animals

The department International Trade of Statistics Netherlands is responsible for the foreign trade statistics. Figures for countries outside the EU are based on the customs declarations. Inside the EU figures are based on the data that enterprises are sending to Statistics Netherlands. Enterprises importing or exporting more than 500.000 Dutch guilder (is about 227.000 Euro) are obliged to send monthly figures. Estimation for enterprises under this threshold and estimation in case of non-response is made. The classification of the goods is based on the SITC. Monthly figures are published divided in 1.250 groups of goods and for 50 (groups of) countries.

Production forecasts (gross indigenous production)

The starting point is the realised GIP in the year before. When forecasting quarter X in year t+1, the base is the GIP in quarter X in year t. There are some "adaptation factors" for correcting the GIP:

The development of the number of mated sows and the number of raised piglets per sow per year.

The development of the number of the monthly exported piglets.

Correction for "not exported" piglets in the quarter before. They will be fattening pigs in the next quarter.

Some special events that will influence the production.

GIP for cattle production is forecast by means of an annual population extrapolation method. This forecast model is based on historical data for the previous years. On the basis of cattle population at the beginning and end of the year and the GIP the annual number of calf births is determined for the previous years. The annual calf rate compared with the initial cow and heifer population will give the calving rate. These rates are used for estimations for the current and coming year. A cattle population and production sheet is drawn up on basis of observation and forecasting. Allowance is made for both the numbers of calves joining the adult population, as either heifers or bulls, and the change in the cow population using coefficients based on data from previous years.

Estimates for the final number of cows and total cattle, which is related to the number of cows gives estimates for the GIP following the calculation: Initial cattle population + calf births - calf losses - GIP = final population.

The mathematical model calculates the GIP for the different categories based on historical data. The results are discussed with experts of the Meat Board and also economic and market factors are checked to adjust any element if necessary.

Especially the figures for foreign trade for live animals are weak elements in the model. Additional data on foreign trade are searched to improve the model for forecasting cattle GIP.

The starting point is the realised GIP in the year before. On the base of the development of the number of animals and the expectation how the market and financial position of the farmers will develop, a correction will made on the GIP.

Poultry statistics

Laying hens total, of which:

younger than 18 weeks;

18 weeks till 20 months;

20 months and older.

Broilers

Parent stock for broilers total, of which:

younger than 18 weeks;

18 weeks and older

ducks for meat production

Turkeys

Other poultry

Monthly slaughtering data for broilers and turkeys are available separately. Monthly slaughtering data for ducks, geese and guinea fowl are not separately available, but the total of these "other poultry" is.

broilers 74%;

old laying hens 70%;

turkeys 80%;

other poultry 70%.

A forecast model on poultry meat production is not available, but could be developed on the basis of incubated eggs or chicks hatched. The basic data for such an econometric model are available.

On the basis of placed young laying hens on farms, a production forecast for eggs is made.