Fishing - pelagic quota cuts 2026: business and regulatory impact assessment

Appendix 1: Methodology

2026 Tonnage and Landed Value Modelling

The global total allowable catch (TAC) figures for North Eastern Atlantic mackerel and North Sea herring used in the analysis assumed an estimated 48% and 22% reduction in mackerel and herring TACs, respectively, compared to the global 2025 figure. The tonnage estimated to be available to the Scottish pelagic fleet was estimated using the 2025 quota share figures the UK, and Scotland, received. The tonnage likely to be caught by the fleet incorporated the fleet’s ability to bank and swap quota year on year, using the relevant quota uptake of Producer Organisations between 2022 and 2024. The final figures estimated and used as the total tonnages available to the fleet in the analysis, were 77,676 tonnes of mackerel, and 44,537 tonnes of herring.

Modelling the estimated tonnage of mackerel and herring landed into Scotland, and abroad, in 2026, was based on the historical fishing activity of the 21 individual vessels within the Scottish pelagic fleet. If a vessel was compliant with a proposed scenario of economic link licence change, then it’s average landing proportions of mackerel and herring into Scotland and abroad, from 2015 to 2024, was assumed to be its landing behaviour in 2026. If a vessel was not compliant with a proposed scenario of economic link licence change (i.e., it’s historical fishing activity did not meet the proposed landing target obligation), its landing proportions were estimated using linear programming, with the set objective of maximising value from mackerel landed abroad. Excel’s in-built ‘Solver’ function was used to model this linear programming, using the Simplex LP solving method. More detail on the tool is available here.

Estimated Wider Economic Impacts

Direct, indirect, and induced impacts were estimated using a mixed exogenous/endogenous variable model of the economy based on the Scottish Government Input-Output (I-O) tables for the fish processing industry. This is the same model that was used to estimate economic impacts in the 2022 BRIA: Amending the economic link licence condition in sea fishing licences Business and Regulatory Impact Assessment.

The model uses an input ratio to convert extra direct inputs (resulting from increased landings into Scotland) into wider supply chain inputs. This is the Indirect Effect, the extra GVA generated through the wider supply chain, as processors demand more inputs (fuel, transport, etc.) to process the higher level of output.

This analysis derived an input ratio of 1.27 applied on any extra landed fish to determine the total inputs used by the fish processing industry. This input ratio was derived by dividing the total intermediate output of the fish processing sector by the primary inputs to determine the number of extra inputs, the primary inputs considered being fishing, aquaculture and the fish processing industry itself (as this would be the secondary processing of fish).

To reduce the risk of overestimating the benefits to the processing sector, the recent high mackerel prices were controlled for by calculating additional inputs per tonne of mackerel in 2018 prices (the final year on which the mixed exogenous/endogenous variables model was based) and using the ONS CPI index to convert to 2025 prices. This was then multiplied by the additional tonnage into Scotland for each scenario to provide the estimate of total input value (including both direct inputs and wider supply chain inputs).

The total input value is then multiplied by the GVA ratio of 1.27 derived from processing data provided by Seafish for Pelagic processors, to find the total output value. The Seafish GVA ratio was calculated by dividing the sum of the GVA and raw material costs for that grouping by the raw material costs. The difference between the total outputs and total inputs is calculated to be the GVA and is expected to include wages, taxes, and gross operating surplus. The total output figure is subsequently used for the upstream calculations.

The rationale for using the GVA ratio obtained by Seafish data rather than using a ratio derived from the I-O table is due to the I-O table including all fish processing. Whereas, Seafish data could instead be split out by industry and showed that aquaculture processing often had a larger GVA ratio than pelagic or demersal 41 processing, which suggests that the I-O table would have overestimated the GVA gains.

The Mixed Exogenous/Endogenous Variables Model

The main task of the model is to estimate the additional output in the economy supported by a given change in the output of the fish processing industry as a result of the domestic fishing industry landing an increased share of its catch in Scotland. The additional output estimated is only that due to backward linkages from the fish processing industry, i.e. that brought about by the industry's increase in purchases of goods and services for intermediate consumption. The additional fish processing output is assumed to be exported or to displace imports.

The standard approach to using an output-driven input-output model is inappropriate here, as it would generate an increase in output of the fishing industry in response to the increase in output of the fish processing sector rather than a shift in use of fishing industry output from exports to intermediate use.

Instead a mixed exogenous/endogenous variables model is used to require that the output of the fishing industry remains unchanged.

The basic equation of input-output analysis is

with the usual solution

where X is the vector of industry gross outputs, A is the direct requirements matrix from the Input-Output table, and F is the vector of final demands for industries’ outputs.

We want to exogenously specify the output of two sectors. Firstly, the fish, fruit, and vegetable processing industry (the narrowest industry classification containing fish processing available in the Input-Output tables), where output is increased by a given amount as a result of processing more fish. Secondly the fishing industry, where output is assumed to be unchanged, as fish that would have been otherwise exported are instead used domestically (note that this implies an assumption of no price difference between selling fish to a domestic processor and selling as an export).

We can rearrange the rows and columns of the matrices and vectors so that the equation above can be rewritten in partitioned form as follows:

with bars indicating the exogenously determined values. This gives us two matrix equations, the second of which is of interest here:

Solving this for X₂ gives

Since we are only interested in changes in X₂ in response to a change in fish processing output, we can set the change in to be zero, and get

the equation used to estimate the changes in industry outputs in response to the increased fish processing activity.

The fish, fruit, and vegetable processing industry also purchases fruit and vegetables from the agriculture industry, and farmed fish from the aquaculture industry. It does not make sense that the fish, fruit, and vegetable processing industry would purchase more of these simply because it increased the quantity of fish it was processing. Therefore the coefficients in A₂₁ corresponding to purchases from the agriculture and aquaculture industries by the fish, fruit, and vegetable processing industry were zeroed out to exclude such effects from the model and prevent them from inflating the results.

Having obtained the vector of output changes in the economy using the equation above, it was then converted into changes in GVA (and FTE) using output to GVA (respectively FTE) ratios from the standard SG Input-Output model.

For further details of the mixed exogenous/endogenous variables approach see:

Johnson, T.G. and Kulshreshtah, S.N. (1982), “Exogenising Agriculture in an Input-Output Model to Estimate Relative Impacts of Different Farm Types”, Western Journal of Agricultural Economics, vol. 7, pp. 187-198.