The Scottish Government

« Previous | Contents | Next »

Listen

Executive Summary

The Single Application Form ( SAF) and the June Agricultural Census ( JAC) are two different forms designed to achieve different objectives for the Scottish Government. Importantly, there is a legal requirement to complete a JAC form under the Agriculture Act 1947, whereas SAF offers a financial reward for compiling forms. Both JAC and SAF have a number of EC statistical reporting requirements.

The JAC covers the whole of the agricultural population of around 50,000 holdings, with a return rate of 70%. The SAF only covers those enterprises which have a history of agricultural support. It has an annual population of 21,000 businesses and around 20,000 are returned by those claiming a Single Farm Payment.

The aim of this study is to assess to what extent information collected from the SAF could be used to meet the statistical requirements of the JAC and how much the JAC could be streamlined or combined with the SAF as a result.

Key findings

Comparison of JAC and SAF forms (2007)

There was a direct match between most of the main Crop questions on the JAC and SAF forms. The JAC contains a bit more detail on the intended date of harvest for ware potatoes and has a couple additional sub-categories for stockfeeding crops.

JAC and SAF both record Grass areas over and under 5 years old, and JAC also enquires about the intended use for mowing or grazing. Both forms ask about Rough Grazing, although the JAC excludes shares in common grazing or seasonally taken in land.

JAC has separate questions on vegetables, fruits and plants grown in glasshouses and under protection, which are not separately asked for in SAF.

The JAC contains more detailed question on Vegetables such as Carrots and Lettuce, although SAF also has some unique categories such as Artichokes and Asparagus. Total vegetable areas should have a close match.

SAF collects more detailed categories of Soft Fruit, although does not separately ask about Strawberry areas which are collected separately on the JAC.

JAC collects more detailed categories of Bulbs, Flowers and Nursery Stock.

JAC collects more detailed information on Pigs, with 13 different categories by type and weight. SAF only distinguishes between indoor and outdoor pigs and also excludes unweaned piglets from the totals.

SAF only enquires about total Poultry, whilst JAC collects information on 11 sub-categories by different types of poultry.

There are 22 categories of Cattle collected on the JAC, compared to just 4 on SAF. JAC collects more detail on the age of animals, whether cows and heifers are in milk or in calf and whether other female cattle are for breeding. SAF excludes cattle aged under 6 months.

There are 13 categories of Sheep collected on JAC, compared with just 3 on SAF. SAF also excludes sheep aged under 6 months old.

For Other Livestock categories, SAF contains more detail on Deer and less on Goats and Horses. SAF excludes other livestock aged less than 6 months, which are included on the JAC.

The JAC collects detailed information on Farm Labour, which is not collected on SAF.

JAC and SAF both collect details on permanent and seasonally rented land. JAC collects more detail on different types of rental agreements.

Comparison of holdings and businesses

For 2007 the coverage of the JAC, by information collected through Field Data Sheets ( FDS) of the SAF was 59% of holdings and 87% of area. The coverage of main holdings was greater than for minor holdings at 81% of main holdings versus 36% of minor holdings and 89% of main holding area versus 61% of minor holding area. This reflects the fact that SAF is returned more often by larger units. It is worth noting that in terms of holdings and area the coverage of JAC by SAF is improving. Staff from the Rural Payments and Inspections Directorate ( RPID) have also indicated that coverage is likely to increase in 2008 since a wider range of holdings will submit SAF as part of making applications for Land Management Contracts ( LMC).

In terms of coverage by farm type we see a wide divergence in the results. For example the SAF coverage by holdings ranges from 21 per cent for horticulture up to 96 per cent for dairy. These results are not unexpected given that specialist horticultural farms are largely outside of the Single Payment Scheme.

In terms of regional coverage, the results show significant regional differences (at the NUTS 4 level), which largely reflects the regional distribution of farm types.

Livestock information within JAC and SAF can only be compared at the business level, as it is not collected at the holding level in SAF. In 2007, 65% of JAC holdings were resolved to 28,365 businesses, with a Business Reference Number ( BRN). The SAF coverage of these was 69% of businesses and 87% of area. However, compared with the whole JAC population, the coverage of area was 78%.

Comparison of data for holdings and businesses common to both the JAC and SAF datasets (2007)

For Crops the range in aggregate differences between the two data sets was from 1% (Spring Barley and Total Cereals) to 362% (Mixed Grains). The larger differences tend to be associated with items with broader definitions e.g. Mixed Grains or for items with smaller areas e.g. Lupins where one or two individual differences can have a big impact overall. For a selection of main crops, aggregate differences showed JAC values greater than SAF for Winter Barley (5%), total Oats (6%), total Oilseed Rape (5%) and total Potatoes (15%) and SAF greater than JAC for Wheat (4%).

The distribution of individual differences for Wheat, Winter Barley, Spring Barley and Oilseed Rape showed that between 37%-42% of main holdings had values within +/-1% on JAC and SAF and between 51%-54% had values within +/-5%. For Seed and Ware Potatoes, there was a sizeable proportion (32%-36%) of holdings showing a positive value on JAC and a zero value on SAF.

The analysis of distributions can be used to identify systematic differences between JAC and SAF, which could be due to factors such as differences in definitions. Another potential difference could be due to estimated values within the JAC dataset, for holdings which have not returned a JAC form. It is recommended that further analysis is undertaken on JAC estimated values.

Analysis of total Grass, shows an aggregate total on SAF 12% (or 134,000 ha) greater than JAC. The distribution of differences shows that 18% of main holdings had values within +/-1% on JAC and SAF, 10% had a SAF value more than 80% greater than JAC and 8% of main holdings had a positive SAF value with zero on JAC. Conversely the aggregate total for Rough Grazing was 11% (or 323,000 ha) greater on JAC than SAF, with 25% of main holdings showing a positive value on JAC and a zero value on SAF. These differences probably reflects the different way that common grazing and seasonal renting of pastures is treated between the two datasets.

The JAC showed an aggregate value for total Vegetables 34% greater than SAF, with 43% of main holdings showing a positive value on JAC and a zero on SAF and 18% of main holdings with JAC and SAF values within +/-1%.

The aggregate difference for total Soft Fruit was 4%, with 18% of main holdings showing a value of +/-1% between JAC and SAF and 35% showing a positive value on JAC and a zero value on SAF.

There are definitional issues between JAC and SAF in the recording of total Bulbs, Flowers and Nursery Stock, with 96% of main holdings reporting a positive value on SAF and a zero value on JAC.

For the main Livestock categories, total Sheep shows the greatest aggregate difference between JAC and SAF of 35% (or 2.5 million). There is a clear peak in the distribution of individual differences, with 41% of JAC businesses showing a value 40%-60% greater than SAF. There are also 14% of businesses with a positive value of SAF and a zero value on JAC.

There pattern is similar for total Cattle, with the JAC aggregate total 12% (or 213,000) greater than SAF and a peak of 26% of businesses showing a value of 20%-40% greater than SAF. There are also 7% of businesses with a positive value of SAF and a zero value on JAC.

The difference in the livestock reference period of March in SAF and June in JAC is likely to be the cause of most of these differences for Sheep and Cattle. In addition, SAF does not collect information on Sheep and Cattle aged under 6 months old.

For total Poultry, SAF shows an aggregate total 9% (or 261,000) greater than JAC. Although 21% of businesses have a JAC and SAF value within +/-1%, there are 24% of businesses with a positive SAF value and corresponding zero value on JAC.

For total Pigs, the aggregated JAC and SAF results are almost identical, with a difference of 0.2% (or 900), however, just 16% of businesses have identical values of +/-1%. Where there is a positive value on both datasets the JAC one tends to be higher, but this is countered by 30% of businesses showing a positive SAF value and a corresponding zero value on JAC. Pig figures for SAF exclude any unweaned piglets.

Examining changes over time (between 2005 and 2007), for crops there is a remarkable degree of consistency in the nature of the differences between JAC and SAF. The year-to-year consistency is maintained for the Grass, hay, straw and silage classes. For livestock Items there is consistency for simple overall categories e.g, Total Pigs, but for classes with more complex or compound definitions there are significant "jumps" in the data series e.g. Beef cows and Heifers.

Comparison of overall totals from JAC and SAF datasets (2007)

The comparison of overall totals from JAC and SAF datasets encompass difference in holdings and businesses common to both JAC and SAF as described above, but it also encompasses differences due to population coverage.

The comparison between JAC and SAF showed that there were 5,084 main holdings and 16,117 minor holdings covered by JAC but not on SAF. Conversely there were 1,055 holdings on SAF which were not on JAC. The following summary quantifies the net effect of additional JAC and SAF population coverage.

For most Crops the impact of additional population coverage was relatively small. For a selection of main crops the difference between JAC and SAF increased by 1% for Wheat, 1% for Winter Barley, 1.4% for Spring Barley, 2.5% for total Oats, 0.7% for total Oilseed Rape and 1.5% for total Potatoes. Larger differences were shown for crops with smaller areas such as Linseed (36%).

Much greater differences were observed for grass and grazing items. The JACRough Grazing estimate increased over SAF by 5.3% (or 521,000 ha) and total Grass increased by 5.9% (or 73,000 ha). This reflects the fact that there are holdings in JAC that are not returning SAF since they have not to date claimed "agricultural" payments, e.g. sporting estates. This is likely to change, however, with SAF used for the administration of LMC and Rural Development Programme ( RDP) measures.

The impact of additional population coverage was small for total Vegetables, with the JAC estimate increasing over SAF by 2.5%. The difference for total Soft Fruit was 8.6%.

For livestock, the greatest impact was for total Poultry, where JAC additional coverage increased by 46.8% (or 2.6 million). There was also a substantial increase for total Pigs of 11.2% (or 50,000). This is not surprising, given that some businesses specialising in poultry and pig production are not eligible to claim for Single Farm Payments ( SFPs).

There was relatively small impact of additional population coverage for total Cattle (0.9%) and total Sheep (1.5%), suggesting that almost all cattle and sheep producers are claiming SFPs.

Integration Issues between JAC and SAF

Having extensively analysed the two datasets the final part of the project comprised a workshop where those involved with the JAC and SAF highlighted their views on key issues surrounding the future possible relationship between the two.

It was recognised that, there is a large degree of overlap, as around 89% of areas are common between datasets. However, the SAF may provide more accuracy, due to its underlying payment and inspection requirements. Hence, a strength would be complementarity between the two data sets, where items missing from one data set could be provided by the other dataset. A similar strength is the willingness to share data among the staff involved in the two exercises, although it was pointed out that currently JAC data can only be used for statistical purposes.

Issues that would need overcoming included:

• reconciling the definitions of ' Business', 'Holdings' and other entities within the SAF and JAC.
• the mismatch in dates on which the data are captured, especially as the SAF and JAC are collected on separate dates and numbers, especially those for livestock, may have changed in-between time.
• possible technical difficulties in exchanging data between the two data models through a lack of common identifiers, i.e. BRNS (Business Reference Numbers).
• the fact that SAF only represents the land on which payment claims are being made and that the SAF population changes on a yearly basis, which would have an impact on any analysis of change within the industry.
• need to rationalise activity codes if the two datasets were merged and possibility of losing detail from the JAC
• SAF and JAC are governed by separate legislation, which may present problems when merging the two.
• The need to ensure that payment dates were met as there would be political consequences if these were not met.

If the above challenges could be overcome there was a view that could lead to some opportunities:

• increased efficiency, as it may lead to reductions in the number of activity codes, as well as farmers only having to complete one form.
• efficiencies from streamlining the processing by the Scottish Government, which would lead to possible cost savings and creating more freedom for staff to take on additional tasks.
• could lead to savings on IT support if there is a move to one dataset. In addition, comparisons with SAF should allow evaluation and improvement of the JAC imputation process.

It was felt that future developments might also offer some opportunities. These include:

• use of web based system as for the SRDP could offer more flexibility and allow more on-line checking of data reducing processing costs.
• The rest of the UK are using BCMS data for cattle instead of the census which has been accepted by the EU. If Scotland could follow this course then this would reduce the amount of information required from the JAC.
• There is an opportunity to simply streamline the SAF and move away from numerous codes measuring specific activities, e.g. no collection of spring barley, just a general set of crop codes.

From the discussion four potential options were considered

1. No change;
2. Reduction in the quantity of information collected within the SAF, which is something can be done easily and presented as a reduction in bureaucracy and may be welcomed by farmers;
3. Reduction in the quantity of information collected within the JAC, which might be possible by using alternative data sources and statistical modelling to meet EC statistical requirements
4. Combining the two forms

It was argued by some at the workshop that items 1 and 4 may not have majority support, but there is an appetite for change and improvement. It was felt that 2 and 3, where the SAF or JAC are simplified, respectively were more possible. The preference for simplification over merger seemed to be driven by the need to focus on securing payment deadlines for farmers and not diverting resources to create awareness of how to deal with the merged forms. However, others argued, that a 'mock-up' form should be made to fully appreciate the implications of merging the two and tested on stakeholders.

« Previous | Contents | Next »