Scotland's Marine Economic Statistics 2019

18. Annex B: Methodology and source data

18.1 Notes about tables

To prevent repetition of notes beneath each table, generic notes are presented in this section. Specific points about individual tables are noted as they arise in the report.

Source data: Scottish Annual Business Statistics (SABS), Office for National Statistics and Marine Scotland. Fishing and Aquaculture figures are taken from analysis of Marine Scotland statistics rather than the SABS figures Table 33 summarises the main data sources for individual topics in this publication.

Table 33: Source data for marine economic sectors

Economic sector and SABS SIC Code	Data sources – economic measures
03.1: Fishing SABS data not used	Seafish Fleet Economic Survey, Marine Scotland Sea Fisheries statistics
03.2: Aquaculture SABS data not used	Marine Scotland Aquaculture Survey statistics: Shellfish Production Survey, and Fish Farm Production Survey. Aquaculture economic and social survey, previously known as Data Collection Framework (DCF) survey.
09.1: Support activities for petroleum and natural gas extraction (extraction is not included in this publication)	SABS
10.2: Processing and preserving of fish, crustaceans and molluscs	SABS Seafish Processing Industry statistics – available but not used Number of Plants – Food Standards Agency
30.1: Building of ships and boats & 33.15: Repair and maintenance of ships and boats	SABS
42.91: Construction of water projects & 52.22: Service activities incidental to water transportation Marine tourism	SABS
50.1: Sea and coastal passenger water transport 50.2: Sea and coastal freight water transport	SABS Transport Scotland statistics
77.34: Renting and leasing of water transport equipment	SABS
Marine tourism	SABS various SIC codes (see Annex A) supplemented by geographic filters to select businesses within 100m of the coast line

Notes:

• Employment figures are head counts (not adjusted to Full Time Equivalents).
• GVA and turnover values are adjusted to 2019 prices.
• Since 2011, support activities for oil and gas were extracted using the SIC code SIC 09.1: Support activities for petroleum and natural gas extraction. However, in 2010 these figures were disclosive and so the wider code SIC 09 Mining support activities was used to provide data. The difference is minimal - see the Oil and Gas Methodology section.
• Totals may not sum due to rounding and disclosure control.

18.2 Scottish Annual Business Statistics

The majority of economic figures in this publication have been taken from the Scottish Annual Business Statistics (SABS) publication. This provides data on a number of economic variables across a range of sectors, based on data from the Annual Business Survey (ABS) conducted by the Office for National Statistics (ONS). The SABS statistics were produced under partnership procedures between ONS and the Scottish Government. These have resulted in an improvement in the quality of the underlying data and consistency in the figures used by ONS and SG.

SABS data is attributed according to business site address e.g. a shop or factory, so data for a large company can be split over more than a single site. Rigorous checks are made to ensure that information relating to individual businesses are not disclosed, either directly or by deduction, in the figures released. In some cases this means that data cannot be presented at smaller geographies such as local authority areas.

To set the individual industry results in context, throughout this publication they have been compared to the Scottish economy as a whole. The national GVA estimate was taken from the Quarterly National Accounts of Scotland. The total employment in Scotland was taken from the Annual population Survey.

SABS data is used to monitor growth sectors in the Growth Sector Statistics Database. Figures from these Marine Economic Statistics are not directly comparable with the food and drink growth sector due to methodological differences.

18.2.1 Sampling methodology

The ABS uses a register of businesses (the Inter Departmental Business Register (IDBR)) to produce an estimate of all businesses that make up the population. The ABS sample is designed as a stratified random sample of UK businesses from the IDBR. The inquiry results are grossed up to the register population, so that they relate to all active UK businesses on the IDBR for the sectors covered.

In 2019, 53% of sampled businesses were included in the final Annual Business Survey (ABS) results UK wide. The average response rate in previous years was approximately 75%. The 2019 rate was lower than usual because the coronavirus (COVID-19) pandemic made it more difficult to contact respondents in order to process survey returns.

For Scotland, this means that the SABS results are based on a lower number of ABS responses than usual. The SABS results for 2019 are based on approximately 6,000 ABS responses from businesses with a presence in Scotland, this is down on the 2018 survey response of around 8,000 responses. This means that the 2019 estimates in this release are subject to more uncertainty than usual. However, we have taken steps to strengthen the estimates for 2019 by incorporating VAT turnover data. More information on this new estimation process is available in the methodology notes. The full set of tables and methodology notes are available under Scottish Annual Business Statistics 2019 heading at: Scottish Annual Business Statistics

18.2.2 Marine economic sectors

Marine economic sectors have been defined as those depending on the marine environment for their output, and where official statistics are available. The SABS data is used along with Marine Scotland’s own statistical publications to gauge each industry’s contribution to the Scottish economy. Where SABS is the sole source of data, the estimates are used directly. Where other sources are used the methodology is introduced in the relevant section of the report and explained more fully in this Methodology section.

Economic values have been adjusted to take inflation into account and are presented at 2019 prices. This means that they are directly comparable in the time series tables.

18.3 Geographic Distribution

The distribution of the whole marine economy, excluding fishing and aquaculture (SIC code 3), by local authority was provided from SABS data. Values for fishing and aquaculture are not derived from SABS and were combined with the SABS extract to produce estimates of all marine economy at local authority level. The fishing values were distributed to local authorities as far as possible, with the residue, which could not be allocated, assigned to an ‘unallocated’ category. Aquaculture values were not available by local authority and therefore these values have been assigned as “Unallocated”.

18.4 Price presentation

All values have been adjusted to 2019 prices using the GDP deflators GDP deflators at market prices making it simpler to compare values across a time series.

18.5 Sea Fisheries Methodology

The GVA and turnover estimates for the fishing sector were provided by Seafish as extracts from their Seafish Industry Authority Fleet Economic Survey. Employment data was from Marine Scotland Sea Fisheries Statistics.

These sources provide a more reliable estimate of economic activity than the SABS figures, initially because Seafish use landings data as the source data for ‘turnover’ and carry out their own financial survey to produce estimates of GVA. The landings data is administrative data that covers the entire population rather than just a sample (as is the case with SABS), while the Seafish financial survey is stratified to produce a representative sample.

Sample sizes

Marine Scotland publishes Scottish Sea Fisheries Statistics as an annual National Statistics publication. The statistics provide a detailed overview of the quantity and value of landings of sea fish and shellfish by Scottish vessels, and landings into Scotland. The Marine Scotland data is a census of all landings by Scottish registered vessels. Information on the Scottish fishing fleet and the number of fishers on Scottish vessels is also presented. Scottish Sea Fisheries Statistics are obtained by data extractions from Scottish and UK databases from data supplied by skippers.

The Seafish Industry Authority carries out economic surveys of the UK fishing fleet, published as annual Economics of the UK fishing fleet publications. The data is also available in the Seafish Fleet Economic Performance Dataset. This contains financial, economic and operation performance indicators for the period 2010-2020. The figures presented in this publication are an extract of economic performance data for all active vessels registered in Scotland.

The Seafish Fleet Economic Survey combines costs and earnings information from vessel accounts provided by vessel owners to the annual Seafish UK Fleet Survey, with official effort, landings and capacity data for Scottish registered vessels from Marine Scotland’s Sea Fisheries Statistics. In late 2020 and early 2021 Seafish collected 354 sets of 2019 financial accounts for the UK fleet (8% of the active UK fleet).

18.5.1 Regional data

Seafish provided figures at local authority level so that a regional distribution of the turnover and GVA can be presented. Data was allocated to the geography based on the location of the registration port of the vessel. Reporting at this level introduces the possibility of disclosing commercially sensitive information. To prevent this happening, all other segmentation categories were aggregated (species type, fishing gear etc.) and outputs where five or fewer vessels were registered were restricted.

18.5.2 Seafish economic analysis methodology

Outputs from the Seafish Fleet Economic Survey provided fishing GVA estimation. Full details of the Seafish methodology are available in the Seafish Fleet Economic Survey, a summary is given here.

The UK fleet is stratified into fleet segments using MMO data^[13] on capacity, effort and landings for each vessel. For the data in this publication, only Scottish registered vessels were included and other segment criteria were suppressed.

Financial accounts were collected for each fleet segment.

Costs and earnings data from vessel accounts were allocated to fleet segments.

Fuel costs and crew costs are calculated differently from the other costs. For crew share, assign a minimum £100 per day, or the actual observed amount, whichever is the higher. For fuel costs, estimate fuel consumption in litres and combine with the average annual red diesel price (excluding duty).

Following calculation of fuel cost and crew share, apply the proportions from the other costs to the official declared fishing income for each vessel to calculate Gross Value Added, operating profit and net profit for each vessel.

UK fleet totals and fleet segment totals and averages are then calculated from the estimates produced for each vessel.

Where there are low sample sizes for a particular segment in a particular year previous years' estimates are taken into consideration.

18.6 Aquaculture methodology

18.6.1 Source data

The quantity and value of fin and shell fish produced on Scottish aquaculture sites is sourced from the Marine Scotland aquaculture production surveys. The aquaculture fish and shellfish production surveys are census surveys, receiving a response from every active fish farm in Scotland.

The financial data for 2018 and earlier is drawn from EU Data Collection Framework (DCF) data. The DCF sets out broad requirements for collecting social, economic and environmental data on aquaculture for the UK. The MMO acts as national correspondent for the data, and CEFAS collected the financial data for the UK. In 2021, Marine Scotland took over the collection of the financial data for 2019 and onwards. This coincided with the UK’s exit from the EU meaning that this collection is no longer covered by the DCF framework. New UK legislation, the UK Fisheries Act 2020 was introduced to replace this, with the aquaculture financial data covered under clause 1. Although the legislation behind this survey has changed, the actual data collected remains unchanged and is comparable to previous years. This survey is now referred to as the Aquaculture economic survey and provides the source financial data for estimating the aquaculture GVA for Scotland.

Although we are legally required to collect the aquaculture economic survey data, aquaculture businesses are not required to respond. The voluntary nature of this survey explains the variation in the number of businesses responding in each year. The table below shows the number of businesses responding in Scotland over the past 8 years.

Species	Count of samples
	2013	2014	2015	2016	2017	2018	2019
Mussels	12	9	5	5	4	3	2
Salmon	14	16	8	8	11	10	7
Trout	3	4	6	5	1	1	2
Total samples	29	29	19	18	16	14	11

Note that although the number of responses are small, the sector is dominated by a small number of very large businesses. Therefore the responses we received covered 74% of the finfish production value and 9% of the shellfish production value.

18.6.2 Calculating GVA using DCF sample data and Marine Scotland survey data

1. The aquaculture economic survey data contains financial returns for Scottish aquaculture businesses, by year and by species type. For this analysis, economic survey data for Scottish production of salmon and trout is used to represent finfish and mussels is used to represent shellfish.

2. Because the Marine Scotland aquaculture production surveys are a census survey, the total production value can be assumed to estimate the value for Scotland. The economic survey sample is scaled to Scotland level using the ratio of the economic survey sample income to the Marine Scotland Aquaculture sample. i.e.

Sample weighting = MS production Survey income / economic survey sample income

3. Estimate GVA by calculating, at the Scotland level, total outputs, total inputs and change in stock value. Note this is different from previous years where change in stock value was not included. However, change in stock value is included in the GVA calculation for the SABS. So, this was included to make them more comparable and to help avoid large year on year changes due to multiyear nature of salmon production.

GVA = Total output – Total input + change in stock value

4. Change in stock value was estimated using the change in stock value from SABS and scaling it up to Scotland level using the ratio of the SABS turnover to the MS production survey income. The SABS change in stock value needed to be scaled up because the turnover was considerably lower than the production survey income so it was underestimating value. It is therefore likely that the change in stock value would also be an underestimate without scaling. i.e.

Estimated change in stock value = change in stock value from SABS × (MS production survey income/SABS turnover)

5. Calculate the GVA to income ratio by

GVA to income ratio= GVA / Total output

This GVA to income ratio can be used to estimate annual GVA where the value of the production is known, e.g. for regional estimates. Economic survey data is available for 2013 to 2019 and a ratio is calculated for each of these years. Table 34 provides the unadjusted GVA and the GVA to income ratio from 2015 onwards. To produce a longer time series, incorporating figures for before 2013, a mean GVA to income ratio was calculated and used to estimate the GVA from the known aquaculture turnover. This is noted in the tables.

Table 34: GVA Ratios for aquaculture production

Mussels

	GVA	GVA to income ratio
2015	5,709,046	0.57
2016	6,295,031	0.54
2017	8,359,512	0.67
2018	5,338,716	0.56
2019	6,755,701	0.84

Salmon

	GVA	GVA to income ratio
2015	68,060,046	0.11
2016	219,884,441	0.29
20171	512,248,168	0.38
20181	373,168,145	0.25
20191	553,461,441	0.49

Trout

	GVA	GVA to income ratio
2015	3,118,759	0.13
2016	2,262,104	0.10

¹ From 2017 Salmon covers all fin fish due to the small number of trout returns.

18.6.3 Comparisons with other data

SABS produces estimates for aquaculture GVA, turnover and employment. These have not been used because the aquaculture production survey statistics are more complete than the SABS surveys. However, estimates for 2015 showed a considerable variance between SABS and estimates based on aquaculture data. This is still that case and may reflect the small number of large aquaculture businesses that dominate the industry and different methodologies used.

18.6.4 Change in calculation of aquaculture GVA

As mentioned in section 18.6.2, the methodology used to calculate aquaculture GVA was revised slightly to make it more comparable with the SABS methodology. This change involved estimating the change in stock value (detailed in section 18.6.2) and including this in the GVA estimate so: GVA=Total output-Total input + change in stock value. This change has been applied to all years of data in this publication.

18.7 Oil & Gas support sector methodology

In this publication estimates for support activities related to oil and gas extraction are presented rather than the extraction itself. This retains consistency with National Accounts Statistics and ABS data where Oil & Gas extraction is normally allocated to a separate ‘Extra Regio’ category rather than allocated to a region within the UK.

ABS data relating to Oil & Gas extraction (SIC 6) is allocated to UK regions (including Scotland) according to the address at which the business is registered - onshore and offshore Oil & Gas extraction and activities are allocated in this way. GVA associated with off-shore activity, under UK regional accounts procedures, is normally allocated to a separate ‘Extra Regio’ category rather than allocated to a region within the UK.

The codes used to produce figures related to oil and gas production have changed from 2010 to 2011. In 2010, data was provided for SIC 09 Mining support service activities. This included both SIC 09.1: Support activities for petroleum and natural gas extraction and SIC 09.9: Support activities for other mining and quarrying. While ‘other mining and quarrying’ activities are not necessarily marine, they were included because publishing only SIC 09.1 at the Scotland level would have been disclosive.

Since 2011, only SIC 9.1 Support activities for petroleum and natural gas extraction have been included. Previous analysis has shown that SIC 09.9 ‘other mining and quarrying’ is a small component of SIC 09 and so the series has been considered comparable and no break in the series is required.

18.8 Fish Processing Methodology

Seafish carries out annual financial surveys of a sample of majority sea fish processors and a biennial census of all UK fish processing businesses. Their Seafood Processing Industry Report presents an overview and analysis of the UK seafood processing industry.

While the Seafish report has the advantage of providing sector specific economic data and does provide information on sea fish processing only (SABS provides all fish processing), it has not been used in this publication because:

The sample size for the financial data is small and not available at local authority level.

Employment is collected biennially and is provided as FTE for Scotland, so it is inconsistent with the headcount figures used in the rest of this publication.

18.9 Water Transport Geographic Distribution Methodology

SABS provides an estimate of the GVA for Scotland for both freight and passenger water transport. These estimates were used as the sole source for economic reporting in this publication.

For the freight transport, Department for Transport (DfT) statistics provide a tables of tonnages of all freight traffic through UK major and minor ports. Using GIS, the Scottish ports were allocated to both Scottish Marine Regions (SMR) and Local Authorities which allowed an analysis of the geographical distribution of tonnage of freight. Passenger transport was not presented by specific ports and so no geographical breakdown was made.

18.10 Marine Tourism Methodology

Marine tourism and recreation has been defined as including “activities which involve travel away from one’s “habitual” place of residence, which have as their host or focus the marine environment and/or the coastal zone”. This is the definition used in the 2015 Scottish Marine Recreation and Tourism Survey.

In this Marine Economic Statistics report, the estimates extract tourism that was identified as ‘marine related’ from all tourism figures as presented in the Scottish Annual Business Statistics data. The approach taken treats all tourism businesses located in postcodes within 100 metres of the coastline as engaging in marine tourism and recreation, or dependent on the marine environment.

The list of SABS codes for relevant tourism and recreation business activities is provided in Annex A. This list aligns with SABS definitions of Sustainable Tourism in their Growth Sector Statistics. The methodology assumes that marine tourism and recreation businesses are located close to the coast. This assumption may exclude relevant activities of tourism and recreation businesses that are located within postcodes that are more than 100 metres from the coastline. The 100 metre threshold was selected to minimise chances of including in the estimates activities of non-marine tourism and recreation businesses that are located along the coast, especially in urban coastal areas such as Aberdeen and Edinburgh.

The March 2021 postcode file was used for the 2018 and 2019 marine tourism data to ensure that the most recent postcode information is used as postcodes and their boundaries can change over time. The April 2018 postcode file continues to be used for the 2017 and earlier years data. The results derived for 2018 using both postcode files were compared to ensure comparability and only very minor and insignificant differences were identified. Therefore all years of data are fully comparable and there is no break in series.

A similar approach to estimating the economic contribution of marine tourism and recreation has been used by other EU member states such as Portugal and the Netherlands. The estimates in this report are therefore comparable with those produced by other countries. In addition, this methodology has been discussed with economists at the Oslo-Paris (OSPAR) Commission Socio-economic Working Group, and there is general agreement across member states on the principles of the approach.

In generating a geographic breakdown of the marine tourism sector, presenting the outputs by local authority was partially disclosive and Scottish Marine Regions (SMR) were selected as the lowest non-disclosive geography that could be used. While SMRs are developed strictly for information related to areas of sea within 12 nm of the coast, they were used in conjunction with the methodology described above to present both marine and land based tourism related to the sea.

18.11 Offshore wind methodology

The Low Carbon and Renewable Energy Economy (LCREE) Survey was designed to provide greater detail on the low carbon and renewable energy economy in the UK. The survey was despatched for the sixth time in 2020, for the reporting year 2019, to a sample of around 24,000 businesses. Information for previous years are subject to revision due to methodology developments, incorporation of additional data or revisions of previous submissions. Typically these revisions only effect the most recent years. The survey collects information on turnover, imports, exports, employment, and acquisitions and disposals of capital assets, for 17 low carbon sectors. These are:

• offshore wind
• onshore wind
• solar photovoltaic
• hydropower
• other renewable electricity
• bioenergy
• alternative fuels
• renewable heat
• renewable combined heat and power
• energy efficient lighting
• other energy efficient products
• energy monitoring, saving or control systems
• low carbon financial and advisory services
• low emission vehicles and infrastructure
• carbon capture and storage
• nuclear
• fuel cells and energy storage

The LCREE figures are survey-based estimates and gather information from a sample rather than the whole population. This means that they are subject to measurable sampling uncertainty, which has an effect on how changes in the estimates should be interpreted. Estimates of the level of uncertainty associated with all figures (coefficients of variation and confidence intervals) reported are presented in the ONS Low carbon and renewable energy economy, UK: 2020 bulletin and datasets to aid interpretation. These uncertainty measures take into account both the variability in the estimate of the proportion of businesses active in the LCREE economy and the variability of the estimate of those active businesses.

In general, changes in the estimates reported are not usually greater than the level that is explainable by sampling variability. This means movements in the estimates should be treated as indicative only. Summary information on the confidence intervals for the 2014 and 2019 offshore wind turnover and FTE estimates are given below:

Table 35: Estimates and confidence intervals for offshore wind turnover and full time equivalent employees, 2014 and 2019

	2014 estimate	2014 lower confidence interval	2014 higher confidence interval	2019 estimate	2019 lower confidence interval	2019 higher confidence interval
Turnover (millions of pounds)	95	12.5	177	602.5	480.5	724.5
Full time equivalent employees (thousands)	0.7	~	1.4	1.6	0.9	2.3

In the case of the estimates above, we can be confident that the turnover in 2019 was significantly higher than in 2014. However, we cannot be confident that the FTE employee estimate was higher in 2019 than in 2014 as the confident intervals overlap. The same applies to the 2018 FTE estimates where we cannot be confident that the real figures in 2019 where any different to the 2018 figures as the confidence intervals overlap.