|
Eurostat Web Site of Eurostat Voice: +352 43011 |
Portugal
The methodology which is described in this chapter is vaild from 2001 onwards.
Livestock surveys
Surveys on the number of pigs
Portugal carries out one pig survey per year, in December. Under the terms of Council Directive 93/23/EEC, Portugal is allowed to dispense with two of the three annual surveys, as it has a pig population of less than three million. The December survey covers pigs only.
The December survey in Portugal is a sample survey with a mean sampling error of 2 %. The last full survey of the pig population was conducted as part of the general survey of agriculture in 1999. The sample includes 53% of the total pig population. The survey covers 2 396 holdings, or 2% of the total.
The pig population is surveyed by selecting a sample based on the register of holdings from the 1999 general survey of agriculture. This frame is updated after each special survey (livestock surveys), after farm structure surveys and, if necessary, with the help of an external survey or administrative sources. For each type of animal, a sample is selected from a total stratified by agricultural region and population size class. For each region, classes are drawn up according to the number of animals.
The survey is conducted in selected regions, namely Entre Douro e Minho, Beira Litoral, Beira Interior, Ribatejo e Oeste and Alentejo.
This sample survey allows extrapolation to the total population. The INE (Instituto Nacional de Estatística - National Statistical Institute) subsequently draws on other sources as a comparative basis for data on regions not included in the survey. These include the structural survey in particular (when available), the annual pig survey (conducted by government bodies, obligatory) and polls of the relevant interest groups.
Surveys are conducted by face-to-face interviews with farmers. The response rate is 100%, any holding failing to provide information being replaced by another with comparable characteristics. Interviews are conducted by the INE in conjunction with the Ministry of Agriculture.
In the medium term (up until 2005), Portugal has no plans to replace statistical surveys of the pig population with administrative data. The way administrative data are currently organised rules out direct access to data. However, as soon as this is possible, it would make sense to draw on such data - they would be used much more effectively, and further surveys of producers could be dispensed with.
Surveys on the number of cattle
Portugal conducts one cattle survey per year, in December. Under the terms of Council Directive 93/24/EEC, Portugal is allowed to dispense with the second annual survey, as it has a bovine population of less than 1.5 million. The December survey covers cattle only.
The December survey in Portugal is a sample survey with a mean sampling error of 1%. The last full survey of the cattle population was conducted as part of the general survey of agriculture in 1999. The sample survey includes 19% of the total cattle population. The survey covers 3 302 holdings, or 2% of the total.
The cattle population is surveyed by selecting a sample based on the register of holdings from the 1999 general survey of agriculture. This frame is updated after each special survey (livestock surveys), after farm structure surveys and, if necessary, with the help of an external survey or administrative sources. For each type of animal, a sample is selected from a total stratified by agricultural region and population size class. For each region, classes are drawn up according to the number of animals.
The survey is carried out in selected regions, namely Entre Douro e Minho, Trás-os-Montes, Beira Litoral, Beira Interior, Ribatejo e Oeste, Alentejo and the Azores.
This sample survey allows extrapolation to the total population. The INE subsequently draws on other sources as a comparative basis for data on regions not included in the survey. These include the structural survey in particular (when available), the annual cattle survey (conducted by government bodies) and polls of the relevant interest groups.
Surveys are conducted by face-to-face interviews with farmers. The response rate is 100%, any holding failing to provide information being replaced by another with comparable characteristics. Interviews are conducted by the INE in conjunction with the Ministry of Agriculture.
In the medium term (up until 2005), Portugal has no plans to replace statistical surveys of the cattle population with administrative data. The way administrative data are currently organised rules out direct access to data. However, as soon as this is possible, it would make sense to draw on such data - they would be used much more effectively, and further surveys of producers could be dispensed with.
Surveys on the number of sheep and goats
Portugal conducts surveys of the sheep and goat populations every December. The surveys are carried out separately and cover the sheep and goat populations only.
The surveys are sample surveys. The last full survey was carried out as part of the 1999 general survey of agriculture. The mean sampling error over the past few years for the sheep and goat surveys has been 2% of the total population of these animals.
The sheep and goat populations are surveyed by selecting a sample based on the register of holdings from the 1999 general survey of agriculture. This frame is updated after each special survey (livestock surveys), after farm structure surveys and, if necessary, with the help of an external survey or administrative sources. For each type of animal, a sample is selected from a total stratified by agricultural region and population size class. For each region, classes are drawn up according to the number of animals.
The surveys cover 9% and 11% of the total sheep and goat populations respectively. 947 sheep holdings and 860 goat holdings are questioned, representing 1% of the total population in each case. The surveys are carried out in selected regions. For sheep, these are Trás-os-Montes, Beira Litoral, Beira Interior, Ribatejo Oeste and Alentejo, and for goats the regions of Entre Douro e Minho, Trás-os-Montes, Beira Litoral, Beira Interior, Ribatejo e Oeste, Alentejo and the Algarve.
This sample survey allows extrapolation to the total population. The INE subsequently draws on other sources as a comparative basis for data on regions not included in the survey. These include the structural survey in particular (when available), the annual sheep and goat survey (conducted by the administration, obligatory) and polls of the relevant interest groups.
Surveys are conducted by face-to-face interviews with farmers. The response rate is 100%, any holding failing to provide information being replaced by another with comparable characteristics. Interviews are conducted by the INE in conjunction with the Ministry of Agriculture.
In the medium term (up until 2005), Portugal has no plans (before 2005) to replace statistical surveys of the sheep and goat population with administrative data. The way administrative data are currently organised rules out direct access to data. However, as soon as this is possible, it would make sense to draw on such data - they would be used much more effectively, and further surveys of producers could be dispensed with.
Notes on the livestock surveys
Surveys of animal populations raise a number of complex problems. This is particularly true of countries such as Portugal, where farming is characterised by a large number of relatively small holdings.
One particularly acute problem is the constant need to update the extensive and extremely complex register. In some cases, comprehensive representative samples must be selected to make up for registers that are not up to date.
Because of the low educational level of farmers, surveys must also be carried out as face-to-face interviews, adding to the logistical and financial burden.
Another complication is the fact that, for odd-numbered years, the data that are needed on livestock populations by holding size must be generated by screening the adjusted findings of the farm structure survey, this study focusing primarily on structural aspects.
Statistics on slaughterings
Portugal compiles monthly statistics on the number and slaughter weight of animals slaughtered in abattoirs for human consumption. These cover pigs (total), calves, heifers, cows, bulls, steers, sheep (total), lambs and goats (total).
Availability of monthly data on slaughterings:
Weight
The monthly census of all public and private abattoirs (some 120 units) is carried out by post by the INE on a country-wide basis. Data are likewise processed by the INE, and are ready on the 19th working day of the month n+2. For slaughterings outside abattoirs (home slaughterings), there are estimated own-consumption coefficients for each type of animal.
Data on external trade in live animals
For data on intra-Community trade, the INE (Division for business statistics, Department for international trade statistics) uses the Intrastat system. Every natural or legal person liable to pay VAT and engaged in the trading of goods between Member States is obliged to take part in a monthly postal survey. Assimilation thresholds are set for individual trade flows. If values fall under these thresholds, data are not recorded. Each year, the INE lays down statistical assimilation thresholds for individual intra-Community trade flows. For 2000, these were PTE 12 million (59 856 €) for imports and PTE 17 million (84 796 €) for exports.
Data on trade with non-member countries are collected using the Single Administrative Document. This is a full survey.
Data preparation (intra-Community trade and non-member countries) is done by the INE only.
Data are available 10 weeks after the reference month for intra-Community trade and six weeks after the reference month for extra-Community trade.
Production forecasts (gross indigenous production)
Forecasts are based on the results of the annual survey of animal stocks carried out in December.
Forecasts for gross indigenous production (GIP) are drawn up over two successive years according to trends in the stocks of certain categories of animals. In other words, the trends in stocks for years n and n+1 are applied to the GIPs calculated using actual data on slaughterings and trade in live animals up until the end of year n+1. This produces a projection of production potential at the time of the survey up until year n+2.
For forecasting purposes, use is also made of other indicators, such as trends in the survey of slaughter cattle, producer prices and international trade in live animals and meat, co-operation taking place with experts from the relevant institutions and interest groups.
The data on international trade used for calculating GIP are the data recorded by the INE on imports/exports (total and intra-Community) of live animals. Account is taken only of the classification codes that relate to the import and export of animals for slaughter.
Poultry statistics
The following data are available in Portugal:
Monthly survey of breeding establishments (Regulation (EEC) No 2782/75) - production of eggs for hatching, production of day-old chicks, capacity of incubators and number of hens capable of laying or not yet capable of laying.
Monthly survey of imports and exports of day-old chicks (Regulation (EEC) No 2782/75) - imports and exports of day-old chicks for breeding or slaughter/laying.
Monthly survey of laying hen establishments - Number of hens capable of laying or not yet capable of laying, plus number of eggs laid in the reference month.
Monthly survey of establishments rearing broilers for meat - Number of arriving hens for meat, number of departing live broilers and number of broilersslaughtered during reference month.
Annual survey of poultry farms (chickens, turkeys, ducks and quails) and interim survey of poultry farms (chickens) - Number of farms by type (meat, eggs or breeding), number of young birds ready or not ready for laying on 30th June and 31st December, annual production of meat, eggs for eating, eggs for hatching and day-old chicks.
Data on poultry slaughterings are not yet collected on a routine basis. It had been planned to expand data collection in this area as part of the 2001 TAPAS action plan ("supply balances"). However, since data are still being analysed, it is not yet possible to supply results on a routine basis.
Use is not made of an econometric model that gives an indication of poultry stocks. The relevant data are recorded by the INE by means of the above-mentioned annual survey and the interim survey of poultry farms.