Back to Ag/Fish Statistics Publications and Data listing
Survey of Tenanted Land in Scotland - 2004
An annual survey of tenanted land in Scotland, primarily used to calculate a value for the Scottish land rental market.
Aim of the survey
To obtain information on the rent paid by Scottish tenants, which can be raised using census data for the total amount of land rented in Scotland to give the gross rent cost for Scotland.
Methodology and sampling
- Postal questionnaires were issued at Martinmas (28th November) 2004. Martinmas is one of the traditional times when rent reviews are carried out. Questionnaires were sent out to a sample of 1013 holding, which is approximately 10% of the population of rented holdings as identified in the June 2004 census. There were 687 usable replies, which gave a response rate of 68%. This is the lower than last year's response rate (76%) and may have been a function of leaving the reminder letter until the new year. In 2005 we will ensure the survey form is issued by 25/11/2005 and the reminder is sent two weeks later, by 09/12/2005. Furthermore, the number of nil or non-useable returns increased to 9% in 2004, from 6% in 2003. Possible reasons for this increase may have been some confusion over the new categories of tenancy and/or an increasing tendency for recipients to feel overburdened with forms.
- The survey was redesigned in 2004 to allow for changes in rent agreements as a result the Agricultural Holdings (Scotland) Bill and will have to be modified next year (2005) to take account of the revised number of farms types, mainly splitting Pigs and Poultry into two distinct farm types when the sample is drawn. Data on the 10 farm types are now available, however when the survey was sent in November only 9 farm types were specified.
- The sample was stratified by LFA type and farm type. LFA denotes Least Favoured Area and is land that is inherently suitable for extensive livestock production but not for the production of cash crops. The alternative is Non-LFA land, which can grow a wide range of agriculture/horticulture crops and where yields are high. There are 9 different farm types that were used to stratify the sample these are: Cereals, General Cropping, Horticulture, Pigs and Poultry, Dairy, Cattle and Sheep (LFA), Cattle and Sheep (Non-LFA), Mixed and Other. A weighted average based on the area of land rented in each category, raised to the total census population gives the estimated rent per hectare figure, which is then used to calculate the estimated gross rent bill.
- In 2003 we changed the sample design. Previously we would over sample bigger holdings so as to reduce the burden on respondents. However in 2003 and 2004 our sample was representative of the Scottish rented sector with no bias towards the larger holdings. We decided to change it as, in 2002, the large holdings were all concentrated in the one area and it resulted in a very low rental value for Cattle and Sheep (LFA) farms. The only problem with this method was that there were no large Non-LFA holdings picked up i.e. those with over 500 hectares. The 2005 sample will therefore be checked to ensure there are adequate numbers of large (500ha+ and 1000ha+) Non-LFA holdings included.
- The sample achieved in this survey provided a fairly even profile of response and is therefore broadly representative of the farms in Scotland. Like any sample survey, the results should not be seen as exact measures, rather as indications of the trend across the wider population of farms in Scotland.
- In 2004 there were 795 questionnaires sent to LFA holdings and 218 sent to Non-LFA holdings.
- The 687 useable replies revealed a total of 779 agreements. After excluding agreements with extreme and non-numeric rental figures we were left with 658 agreements to be analysed.
- The number of agreements falling into each type of tenancy category are shown in Table 1, below. The removal of the Crofting Tenancy category from this year's survey coincided with an increase in the proportions of agreements within the 'Full Agricultural Tenancy - Tenant' and 'Other' categories. The new Limited Duration Tenancy (LDT) and Short Limited Duration Tenancy (SLDT) categories were selected by a relatively small proportion of respondents. Furthermore, the returns suggest that not all respondents understood the new categories they chose. Two of those who selected an SLDT believed the term length of the agreement to be indefinite and one respondent indicated the term of their SLDT is 15 years.
Table 1 - agreement categories
|Types of agreement||No.||%|
|Full Agricultural Tenancy - Tenant||463||70.4|
|Full Agricultural Tenancy - Limited Partnership||96||14.6|
|LDT converted from Full Agricultural Tenancy||2||0.3|
|Contract let or management agreement||7||1.1|
- The average rent per hectare for LFA land was found to be £17.53 in 2004; this is an increase of 5.7% on the 2003 figure of £16.58. The average rent per hectare for Non-LFA in 2004 was £112.94, which is an increase of 2.0% on last year's estimate of £110.74. Overall in 2004, the average rent per hectare was £31.10, which is a 6.2% rise from the 2003 estimate of £29.29 (note the higher percentage rise overall is a function of the weighted average calculation; the area of rented LFA land decreased and the area of rented non-LFA land increased from 2003 to 2004).
Table 2 - rent per hectare by farm type for 2003 and 2004
|Farm Type||2003||2004||% change|
|Cattle & Sheep (LFA)||23.89||26.91||12.6|
|Cattle & Sheep (non-LFA)||105.74||89.14||-15.7|
(results for horticulture, pigs and poultry are excluded from this table due to small sample sizes)
- The results in table 2 are not weighted. The figures for the Cattle+Sheep (Non LFA) category seem to have fallen substantially, however, the 2004 survey saw a particularly low number of responses (8 from 19 in 2004, 18 from 23 in 2003) from this category of farm. This trend will be monitored in the 2005 survey. If the trend of low response continues we can attempt to rectify in the future by over sampling this category. The average result for the General Cropping farm type may also have been affected by a lower response than last year. Historically, holdings in the Cattle+Sheep (LFA) and Other categories tend to have lower rents than the Cereals, General Cropping, Dairy, Cattle+Sheep (Non-LFA) and Mixed farm types
Chart 1- rent per hectare by farm type
- The amount of land rented and the land quality heavily influence the rent paid per hectare.
Table 3 - rent per hectare by LFA type and rented land size group for 2004
|Size Group||Rent per ha||Rent per ha|
(no response from Non_LFA 500-<1000 and 1000+ categories)
- In the 2004 survey we asked for information on seasonal rents, from the 687 analysed replies there were 138 seasonal agreements identified. The results revealed that the average rent per hectare of seasonal land was £61.13, this is a weighted average based only on the census distribution of area; there is no adjustment for land quality or farm type as this information is not available from the census or the survey.
- Rented land in Scotland has been declining year on year since 1992; chart 2 shows a time series of the area of tenanted land in Scotland from 1987 to 2004.
- The annual average reduction in tenanted land area since 1993 is 2.3%. From 2003 to 2004 the amount of tenanted land reduced by 0.7%, from 1,660,823 to 1,648,992 (hectares). In no other year since 1993 has the fall been less than 1%.
Chart 2 - rented land (1987-2004)
- Despite a reduction in rented land in 2004 compared to 2003, the increased rent per hectare value has resulted in a rise in the gross rent total to £51.3 million in 2004 compared with £48.6 million in 2003. Chart 3 illustrates the changes in the total gross rent value.
Chart 3 - Gross rent (1998-2004)
- The Gross Rent and Total Income From Farming (TIFF) series are compared below.
Table 4 - TIFF and Gross Rent (1998 - 2004)
|Year||TIFF||TIFF index||Gross Rent||Gross Rent index|
|(nominal £m)||(2000=100)||(nominal £m)||(2000=100)|
Chart 4 - TIFF and Gross Rent (1998-2004)
Back to top
- Analysis indicates a weak negative relationship between TIFF and Gross Rent, however the model derived from this analysis does not provide a statistically significant explanation of the relationship between these two series.