Methodology changes and corrections

Scottish farm business income estimates: 2024-25, published 26 March 2026

This release includes final estimates for 2023-24 and initial estimates for 2024-25.

Farm business income estimates for 2023-24 and 2024-25 were impacted by Basis Period Reform (BPR), improvements to sample coverage and improvements made to data processing.

These methdology changes have no impact on the comparability of 2023-24 and 2024-25 estimates with the timeseries, dating back to 2012-13. The size of the impact on estimates remains small compared with the margin of uncertainty around survey results.

Basis Period Reform

Under BPR, the accounting year of farm businesses is aligning with the financial year. As a farm transitions, its previous year closing valuation is re-estimated to align with the financial year. New estimates are based on the previous valuation and available data.

In 2024-25, 53 farm accounts (13%) were impacted by BPR. In 2023-24, 110 farms (28%) were impacted by BPR.

More information is available in Basis Period Reform.

Correction to 2023-24 estimates related to Basis Period Reform

Estimates for 2023-24 have been corrected in the 2024-25 publication. A notice on the 2023-24 publication directs users to corrected results.

An error was identified for some 2023-24 estimates of farm business income. The error occurred due to changes in methodology related to Basis Period Reform (BPR). 

Annual farm business income estimates are based on the fully reconciliated opening and closing valuations of farm businesses during an accounting year. Under BPR, the accounting year end of farms is moved to align with the end of the financial year. An error occurred in 2023-24, where due to the change in year end, potato outputs were missed from the new 31 March closing valuation of some general cropping farms.

As a result, some average estimates for the whole sample (“all farms”) and general cropping farms were impacted. A list of estimates impacted by this data error and the workbook where these estimates were first reported on 3 April 2025 is available.     

Impact of sample coverage improvements on final estimates

As we continue to improve sample representation, more smaller farms were recruited in 2024-25. Incorporating a back-year of data for these newly recruited farms led to larger differences than usual between the initial and final estimates for 2023-24. This is because smaller farms typically report lower incomes than larger farms. The size of the impact of changing sample composition is small compared with the overall uncertainty in 2023-24 results.

Estimates for each survey year are finalised the following year. For example, initial estimates for 2023-24 were published in April 2025. Final estimates for 2023-24 were published in March 2026.

Final estimates can include late data returns and the addition of a “back-year” of data for new farms joining the sample. For example, if a farm enters the sample in 2024-25, it can also provide data for 2023-24 if available. This additional information is included in the final estimates for 2023-24.

This approach is consistent with the methodology used across farm business income statistics. Typically, only small differences arise between initial and final estimates for any given year. However, changes in the sample can have a noticeable impact on the level of variability between these two sets of estimates.

In 2024-25 more small Less Favoured Area (LFA) sheep and lowland cattle and sheep farms joined the sample. The back-year farm business income 2023-24 estimates for these farms were, on average, lower than initial 2023-24 estimates for LFA sheep and lowland cattle and sheep farms already in the sample. This resulted in lower average income estimates in the final 2023-24 results compared with the initial 2023-24 results.

Where changes are strongly attributed to changes in the sample representation, these are reported on alongside the official statistics.

For example, the official statistics publication released on 26 March 2026 notes that:

“Sample changes in lowland cattle and sheep also affect these results and cannot be fully separated from year-on-year change. Farms exiting the survey in 2023-24 had higher milk outputs compared to the remaining sample. Incoming farms in 2024-25 had higher diversification income (rental income and food processing) compared to the previous sample.”

The final 2023-24 income estimates for lowland cattle and sheep farms were also affected by improvements to the analytical processes used to produce the estimates.

More information about the FBS sample is available in the Farm business income methodology and Accuracy and Reliability sections.

Improvements to analytical processes

Ongoing improvements are being made to the analytical processes used to produce income estimates and to weight results to the June Agricultural Census (JAC) population. These developments form part of a wider replacement of existing methods with reproducible analytical pipelines. The transformation aims to improve both the quality and accessibility of outputs.

As part of the improvement to data processing, we are also addressing minor inconsistencies within the current methodology. For the weighting process specifically, updates have been made to underlying variables to ensure closer alignment with recent changes in the JAC dataset. Updates also include corrections to the number of JAC holdings included in the weighting.  

Corrections to the weighting had a negligible impact for all farm types, compared to the margin of uncertainty around survey results. However, due to the relatively small sample size of lowland cattle and sheep farms, updates to the weighting resulted in larger shifts in final estimates from initial estimates, compared to other farm types.

The weighting process incorporates the physical characteristics of farms in the sample (such as farm type, size, and land area). This means the scale of the change is affected by any differences between farms included in the initial and final samples.

In 2023-24, the final sample included more smaller lowland cattle and sheep farms. As these farms typically report lower incomes, the updated weighting contributed to even lower final estimates relative to the initial figures.

Final 2023-24 income estimates for lowland cattle and sheep farms were 32% (£1,300) lower than initial estimates with the updated weighting increasing the difference to 46% (£1,900). These differences are still small compared with the overall uncertainty in the 2023‑24 results.

Weighting updates do not affect the comparability of data over time. Only data for 2023-24 and 2024-25 have been updated with the improved weighting method. Data back to 2012-13 have not been revised and the methodology remains consistent with recent data.

More information about the weighting method is available in Weighting.