Scottish farm business income: annual estimates: Methodology

Methodology for Scottish farm business income estimate publications


Farm business income methodology

Estimates of farm income in Scotland come from the Scottish Farm Business Survey (FBS). The main figure reported is farm business income (FBI). The FBS is based on an annual survey of approximately 400 farms.

Data are collected by SAC Consulting (SACC) on behalf of the Scottish Government. SACC recruit a sample of farms and collect data through farm visits and detailed examination of accounts and paperwork. When complete, anonymised data is passed to the Scottish Government for analysis and publication.

Currently, around 400 fully-reconciled farm accounts are compiled each year. These are constructed using information supplied by co-operating farmers and the data quality is very high.

SACC collects detailed financial and economic information for the farm business on outputs, inputs, income and balance sheets. The financial information is collected to observe the overall performance of the farm business for a particular year, and to construct a full profit-and-loss account and a balance sheet.

Some physical information, such as crop areas and stock numbers, is also collected. The physical data is used primarily to classify the farm according to its type and tenure.

Data for each farm is validated against a comprehensive set of quality assurance checks.

Some information on non-cash items, such as input of family labour, is estimated to complete the economic picture of the business.

Information is also collected, where possible, for the farmer and spouse on their non-farming income. Non-farming income means income from other employment or self-employment, investments, pensions, and social payments.

The survey is not carried out on a calendar-year basis, but is based on farms’ financial years. The spread of closing valuation dates goes from the autumn of one year to the spring of the next, which means that some of the 2022-23 accounts relate to the 2021-22 winter whilst others relate to that of 2022-23. Although the accounting year-ends for individual businesses in the sample vary, they all centre on the same cropping period. For example, the 2022-23 accounts all centre on the 2022 production and support payments year.

The FBS results are obtained from a sample of farms that are stratified by farm type and economic size. The coverage of the survey is restricted to those farms which have considerable economic activity (at least €25,000 of output, equivalent to approximately £20,000) and are not considered as spare-time farms (have a Standard Labour Requirement [1] of more than 0.5).

The sample excludes farm types not in receipt of support payments. As such, the survey does not include farms predominantly engaged in horticulture, poultry, egg production or pig production.

Data from the June Agricultural Census is used to weight the farm business income estimates. The FBS is representative of around 10,500 farms in Scotland, which in 2022-23 is around 19% of all farms included in the June Agricultural Census. These farms cover 63% of Scotland's agricultural land, account for 45% of the Scottish farming labour force, and produce 73% of total standard output. The survey is designed for coverage of commercial sized farms, and represents 95% of standard output from these farm types.

Classification of Farms

The classification is based on detailed sub-types as defined in the European Commission (EC) farm typology [2], which have been grouped together where required to give the types shown below.

The classification is based on the relative importance of the various crop and livestock enterprises on each farm assessed in terms of standard output. The method of classifying each farm is to multiply the area of each crop (other than forage) and the average number of each category of livestock by the appropriate standard output. The largest source of output determines the type of farm. The list below defines the main types that are reported in the Farm Business Survey:

  • Specialist Sheep (LFA) - Farms in less-favoured areas with more than two-thirds of their total standard output coming from sheep.
  • Specialist Beef (LFA) - Farms in less-favoured areas with more than two-thirds of their total standard output coming from cattle.
  • Cattle and Sheep (LFA) - Farms in less-favoured areas with more than two-thirds of their total standard output coming from sheep and beef cattle together.
  • Cereals - Farms with more than two-thirds of their total standard output coming from cereals and oilseeds.
  • General Cropping - Other farms with more than two-thirds of their total standard output coming from all crops.
  • Dairy - Farms with more than two-thirds of their total standard output coming from dairy cows.
  • Lowland livestock - Farms not in less-favoured areas, with more than two-thirds of their total standard output coming from sheep and beef cattle.
  • Mixed - Farms with no enterprise contributing more than two-thirds of their total standard output.

Sampling

The sampling strategy of the FBS is based on a stratified simple random sample and is effectively designed as a panel survey with little change in the membership of the sample between years. The sampling frame for the survey is the Scottish Agricultural Census, according to the specific farm type and standard output size requirements of the FBS sample.

An important feature of the survey is the measurement of changes in farm incomes and in incomes from diversified activities for particular types of farm for at least two years. To achieve this, it is necessary to maintain farms in the sample surveyed over a number of years. Once recruited, the farm may stay in the sample for an unlimited time period. The involvement of farms in the FBS is entirely voluntary.

If farms drop out of the survey, replacements are selected depending on which farm types, economic output and district of Scotland are required to achieve a sample which is representative of the population of farms in Scotland.

Weighting

Due to the nature of survey samples, the individuals being sampled may not directly reflect the population that those individuals are being sampled from. This can cause under- or over‑representation of particular farm characteristics. For example, there may be a high proportion of dairy farms that are owner-occupied in Scotland yet, due to location or the voluntary nature of the survey, this high proportion may not be fully represented. Weightings are applied to the data to account for this sort of issue.

To account for the potentially unrepresentative nature of the sample a calibration weighting method has been applied, making the weighted averages more representative of the population. Data has been weighted back to 2012-13. Comparisons between weighted data prior to this time should not be made because the weighting methodologies are different.

Weightings are based on the June Census distribution of agricultural holdings in Scotland and are broken down by eight farm types and three farm tenures.

With a calibrated weighting method, the calibration variables are summed to equal the estimated population totals. The calibration variables used are:

  • type of farm
  • tenure type of farm
  • area of barley
  • area of oats
  • area of potatoes
  • area of wheat
  • rented area of land
  • total area of farm
  • number of dairy cows
  • number of beef cows
  • number of ewes

This weight, when applied to each farm, represents the number of times that farm data must be replicated to ‘represent’ farms not selected for in the sample. By weighting in this calibrated way, bias that may occur due to common problems related to survey data is reduced. A more precise estimate for the variables being analysed, even when the groups sampled do not directly reflect the population, is therefore provided.

Back to top