Research and analysis

Engineering Biology, Quantum Technology and Robotics and Autonomous Systems: Company Performance and R&D Tax Credit Estimates

Published 25 September 2025

1. Summary

This publication provides estimates based on HMRC data to measure the size and economic performance of 3 tech sectors in the UK:

• Engineering Biology (‘Eng Bio’)
• Quantum Technology (‘Quantum’)
• Robotics and Autonomous Systems (‘RAS’)

The analysis shown here uses DSIT’s novel sector identification methodology and matches company lists to HMRC tax data, to provide indicators of performance (company performance estimates) and insights into companies’ research and development investment (R&D tax credits estimates) over time.

Historically, DSIT has relied on commercial databases, but these provide incomplete data with poor coverage for key variables. The new approach shown here attempts to overcome this limitation. The estimates provided here are more reliable than those based on externally sources because HMRC tax data are less prone to reporting biases or inaccuracies.

While this new approach is deemed an improvement on previous approaches, it is not without caveats. Several steps were taken to convert administrative tax data into meaningful information, such as matching data across tax systems, annualising reported data into financial periods, and making certain informed assumptions. As a result, the statistics presented are estimates. Furthermore, the estimates rely on administrative tax data, which is subject to frequent revision, meaning they should be interpreted with caution.

The analysis uses tax data from Corporation Tax (CT) and Pay As You Earn (PAYE) and insights are aggregated to avoid being disclosive. The fundamental structural differences between these taxes mean that financial years refer to different periods of economic activity by different companies and are not directly comparable. Furthermore, the company lists which identify the 3 tech sectors are recent snapshots, taken at different times, and variation in the sector population over time is not captured in the time series shown. These timing issues are explained in detail in section 6.5 in the Annex.

Nevertheless, the methodology used to produce these estimates offers valuable new insights into the economic performance and R&D investment of tech sector companies and will help to inform policy development.

2. Headline findings

Headline findings are broken down into company performance estimates and R&D tax credits, due to the different sources used and the nature of the statistics. Where tables show “N/A” this indicates suppression to prevent disclosure.

2.1 Company Performance Estimates

Company performance estimates measure turnover, employees, wage expenditure, profit and gross value added (GVA). The methods used to elicit these findings are explained in section 6.6. in the Annex.

Table 1 and Table 2 show figures for companies with accounting periods (APs) ending in the financial year 2022-2023 for estimates drawn from Corporation Tax (CT) data, and employers for the tax year from 6 April 2022 to 5 April 2023 for estimates drawn from Pay As You Earn (PAYE) data. GVA is estimated using a bespoke calculation using variables from both tax data sources.

Table 1: Large companies: company performance headline estimates, APs ending in 2022-2023 for CT variables and tax year 2022-2023 for PAYE variables

Large Company estimates for 2022-2023	Eng Bio	Quantum	RAS
Number of active companies	56	10	37
Total Turnover (£m)	34,340	22,050	9,250
Partially-gross trading profit (£m)	2,210	N/A	160
Employee count	32,700	34,220	47,970
Median employee count	250	2,680	390
Total Expenditure on Wages (£m)	2,690	2,140	2,140
Average Employee Pay (£k)	72	52	33
GVA (£m)	5,010	N/A	1,890

Table 2: SMEs: company performance headline estimates, APs ending in 2022-2023 for CT variables and tax year 2022-2023 for PAYE variables

SME estimates for 2022-2023	Eng Bio	Quantum	RAS
Number of active companies	1,448	76	2,355
Total Turnover (£m)	2,580	220	3,870
Partially-gross trading profit (£m)	-3,340	-90	-380
Employee count	17,790	1,680	23,420
Median employee count	8	13	6
Total Expenditure on Wages (£m)	1,070	100	1,080
Average Employee Pay (£k)	48	51	37
GVA (£m)	-1,760	10	760

This analysis estimated that, in some periods, the GVA of these sectors was negative. It is therefore proposed that, for these R&D sectors, GVA may not be an appropriate indicator of value because:

• We would fully expect smaller, early-stage companies with capital-intensive investment needs and private equity-backing to show negative GVA.
• These tech sectors are nascent and working at the edge of new technology and GVA may not therefore appropriate as a lead indicator at this stage. GVA may not represent value when considering the R&D activity of these companies because accounting standards have not evolved to recognise investments in intangible assets as being equivalent to investments in physical capital and, for that reason, they cannot be easily included in GVA calculations.

Further work is required to assess whether traditional GVA measures are appropriate for R&D sectors.

2.2 R&D Tax Credits

R&D tax credit estimates for these sectors are outlined in table 3 below and detailed further in this publication. All R&D tax credit estimates are drawn from Corporation Tax data.

Table 3: R&D Tax Credit Headline Estimates, UK, 2021-2022

R&D Tax Credit Estimates, accounting periods ending in 2021-2022	Eng Bio	Quantum	RAS
Number of active companies	1,420	78	2,281
Number (%) of active companies claiming R&D tax credits	538 (38%)	53 (68%)	598 (26%)
Number (%) of active companies claiming R&D tax credits under the SME scheme	274 (19%)	N/A	377 (17%)
Number (%) of active companies claiming R&D tax credits under the Research and Development Expenditure Credit (RDEC) scheme	131 (9%)	N/A	125 (5%)
Number (%) of active companies claiming R&D tax credits under both schemes	133 (9%)	N/A	96 (4%)
Total R&D qualifying expenditure, large companies (£m)	1,170	N/A	530
Total R&D qualifying expenditure, SMEs (£m)	980	N/A	350

3. Background

DSIT’s Technologies, Growth & Security analysis team has developed company lists through natural language processing, which identify how many businesses there are in each sector, where they are, and what they do. However, due to the incompleteness of commercially accessible financial datasets, they were unable to accurately quantify key metrics of interest such as turnover, employment, and growth.

To overcome this, DSIT explored using HMRC tax data. This report presents aggregate analysis of the general company performance metrics of interest, produced by matching DSIT’s company list onto HMRC’s Companies Summary across Taxes (CSAT) database and the Companies Database (CDB), which are more complete and provide variables better suited to the analysis, which are drawn from Corporation Tax (CT) and Pay As You Earn (PAYE) tax return data. CSAT also includes Value Added Tax (VAT) data, but these variables are not used here.

In addition, HMRC’s data on R&D tax credit claims has also been explored. Claims data comes from Corporation Tax return data but is stored separately from CSAT and CDB and thus required a separate data matching process. Like employment, GVA and turnover, this information will help inform policymaking at DSIT. For example, the aggregate amount of R&D rebate used gives an indication of the extent to which the sector is financially supported by government.

4. Methodology

4.1 Sector identification

4.1.1 Process to developing company lists

DSIT analytical and policy colleagues, alongside subject matter experts, used the Data City^{[footnote 1]}’s supervised machine learning tool to develop company lists for Eng Bio, Quantum and RAS sectors. This process produced comprehensive lists of companies operating in each sector across the UK at a specific point in time for each sector. Within each sector, aligned to taxonomies developed by sector teams and cross-government consultations, the process involved developing company lists for sub-sectors within each overall sector. These taxonomies and descriptions can be found in section 6.6 in the Annex.

4.1.2 Limitations of company lists

The process of using the Data City and web-scraping machine learning means that tech sector lists may be biased towards companies manufacturing or developing specific technologies. Whilst many companies work across tech areas, the process means that company lists may not include larger conglomerates who also manufacture or integrate technologies, thus including more dedicated companies. Whilst analysis is undertaken separating the large companies from SMEs, it is likely that the large companies category analysed here does not include some relevant diversified companies. This ultimately impacts sector estimates and must be recognised when interpreting these findings.

The sector lists were produced in 2023 and 2024, which creates a single snapshot of active companies. This means that sector estimates can be produced and aggregated for this company list historically, but this does not provide figures for companies who became inactive within the timeframe and were not captured in the company list. For example, a tech company who became inactive in 2022 will not be measured in these estimates, as the ‘snapshot’ 2023 company list would not include this company. Instead, estimates were produced for all companies still active, including ‘new’ companies established within the reporting period. Recognising that new companies become active over time (an increase in the number of companies reporting data, rather than the variable of interest increasing for the same companies) this publication provides figures for a ‘cohort’, which is a subset of all companies in each sector that were active in 2018-2019; the first accounting period covered in this publication.

4.2 Data Matching

4.2.1 Data Inputs

DSIT provided HMRC with lists of company reference numbers (CRNs) for the matching process for each sector. The companies identified include subsidiaries; however, some company information may be recorded for different subsidiaries, and tax data is reported to HMRC at individual company rather than group level (for Corporation Tax; note that PAYE information is not necessarily reported at the same level as CT returns). Hence, all are matched into the analysis here for completeness.

The aim was to produce the most accurate estimates of key metrics for each sector. Where we report results on the number of companies per year, this will be at the company rather than group level. This caveat must be kept in mind when quantifying the growth in the number of companies in the sector.

DSIT analysts used an agile approach to collate several other variables. For general company performance estimates (i.e. employees, wages, income etc):

• Sub-sector flags indicating which companies within each sector fell into which sub-sectors. This is taken directly from the sub-sector lists produced for the Data City company lists.
• For Eng Bio and RAS, a flag indicating whether companies were equity funded, using Beauhurst^{[footnote 2]} data and Data City data .
• For Eng Bio, a flag describing if the company is deemed in the supply chain, application, or both. The breakdown of Eng Bio sub-sectors is explained further in section 6.6 in the Annex.

For R&D tax credit analyses:

• A flag indicating which companies have received funding from Innovate UK (‘iUK’) grants, taken from Data City data.
• For RAS and Eng Bio companies, a flag indicating which companies received a grant from other grant providers, taken from Data City data.
• A flag indicating which companies have received funding from the British Business Bank. Company lists were matched with the British Business Bank’s Management Information data in October 2024.

4.2.2 Interpreting HMRC’s Tax Data

All HMRC datasets used in this analysis include data from Corporation Tax returns. The convention for presenting Corporation Tax data over time is to break it down into financial years based on the end dates of company accounting periods rather than the points in time at which the company’s economic activity took place. This is because there is no standard ‘tax year’ for Corporation Tax and companies may have accounting periods beginning and ending at any point in time (but not exceeding a year in length). This presentational convention is known as the “CT600 accounting basis”.

For example, turnover or R&D expenditure pertaining to a company with an accounting period beginning in May 2017 and ending in April 2018 will be included in the 2018-2019 financial year on this accounting basis, and so will the turnover and expenditure for a company with an April 2018 to March 2019 accounting period. In other words, on the CT600 accounting basis, activity may appear to take place in the same year even though for some companies the actual timing may differ by up to a year depending on their accounting period end dates. This should be borne in mind when interpreting the findings in this report.

For Corporation Tax (and VAT data) in CSAT, the variables are also annualised (see section 6.1 in the Annex for more information on this process). This is not the case for the R&D variables, which are drawn from Corporation Tax data but have not been annualised.

The presentational convention for Corporation Tax data (whether sourced from CSAT or the Companies Database) means that Corporation Tax variables are presented on a different basis to the CSAT employment variables, which are drawn from PAYE data and are shown on a ‘tax year’ basis. For PAYE variables (namely those relating to employment), the ‘tax year’ runs from 6 April to 5 April of the following year.

4.2.3 Matching Processes

HMRC carried out 2 complimentary matching processes – one aimed at producing general company performance estimates and the other aimed at capturing R&D tax credits (reliefs).

4.2.3.1 Company Performance Estimates

HMRC first matched the company lists to the CSAT database via Company Reference Number (CRN); we prioritise CSAT as the main source of data for the analysis. CSAT includes all active companies from Companies House, and those which have ceased since April 2015, and contains information from multiple tax heads such as PAYE, VAT and Corporation Tax. More information on the CSAT database is provided in the Annex, including on the company to PAYE scheme matching.

The Eng Bio company list was produced in October 2023, whereas the Quantum list was finalised in April 2024 and the RAS list was finalised in September 2024. Using the CSAT database as of July 2024, HMRC matched over 99% of CRNs in the Eng Bio and RAS sectors to CSAT and 100% of the CRNs in the Quantum sector.

After matching CRNs with HMRC databases, data was extracted for the matched CRNs for 2018-2019 to 2022-2023, which refer to the financial years in which company accounting periods ended (for CT variables) or the PAYE tax years.

All company information used in this analysis is presented at the CRN level, no attempt has been made to aggregate to the group level.

Further detail about the variables used in the analysis, including their sources and derivations, is presented in the Annex.

4.2.3.2 Variable Coverage

Not all companies had the variables of interest present in the databases. Detailed figures on the coverage of the data for each variable are presented in the Annex.

To measure the coverage, we accounted for the increasing number of companies per year (as new companies become active) by selecting only the active companies with accounting periods ending in that financial year (or reporting in the PAYE tax year). We define an active company as one that has not ceased, and is not dormant, in liquidation, or in administration. We then quantified the percentage of these active companies that we have reported data for.

Overall, the coverage amongst active companies varied across sectors and years as follows:

• Annual Turnover: around 70-80% - turnover is not a mandatory variable on the CT return;
• Partially Gross Trading Profit (PGTP): around 80-90% - while the components of this variable are mandatory on the CT return, not all active companies will be trading, so not all will have trading profits to report;
• Employee Variables: around 60-80% - not all companies will have employees and be registered for PAYE;
• GVA: around 55-75% - this is a composite variable whose coverage depends on the intersection in its components’ coverage. GVA has only been estimated for companies where there has been both PGTP and Wages variable coverage.

More detail is presented in the Annex about variable coverage.

DSIT assessed the coverage of alternative data providers and found that the coverage of these variables was typically under 30%, where providers produced estimates for incomplete data. Thus, while HMRC’s tax data coverage is imperfect, it nevertheless leads to more robust estimates of the key company performance metrics of interest.

4.2.3.3 R&D Tax credits estimates

Of the company CRNs provided by DSIT for the Eng Bio sector, around 30% of CRNs were successfully matched to HMRC’s R&D claims dataset on average per financial year. The equivalent proportions were around 55% for the Quantum sector and around 20% in the RAS sector. Those without a match in a given financial year can be assumed not to have made any R&D claims in accounting periods ending in that year.

Other HMRC data sources, namely CSAT and the Companies Database (CDB) were used to identify each company as active/inactive and large/SME.

We define an active company as one that has not ceased, and is not dormant, in liquidation, or in administration. Companies NOT classed as active are therefore omitted from the R&D analysis as they are deemed to be inactive for the purposes of this analysis.

Like many of the company performance estimates, the R&D data were sourced from Corporation Tax returns for accounting periods ending in the specified financial years, so are presented on the CT600 accounting basis, as described above.

During the period shown in this analysis , there were 2 Corporation Tax Credit (CTC) schemes under which companies may claim tax credits subject to their eligibility: the Research and Development Expenditure Credit (RDEC) and Small or Medium-sized Enterprise (SME) schemes. Irrespective of the scheme, companies receive tax credits (reliefs) through 1 of 2 mechanisms – Corporation Tax liability deductions or directly payable credit. The tax credit (relief) amount is determined by the company’s qualifying expenditure.

Different CRNs (companies) may belong to the same group structure. While each individual company is required to submit its own Corporation Tax return, companies in group structures can share their profits and losses among themselves to make tax savings where applicable. For this reason, groups may choose to limit R&D activity to certain companies within the group, even though all companies in the group structure will ultimately benefit from the return on the investment. Companies sharing the same company website URL were assumed to belong to a group structure for the purpose of group-level analysis, which aims to give a more accurate representation of the proportions of sectors that benefit from R&D tax credits.

4.2.4 Classifying Large and SME Companies

Preliminary investigations into the CSAT data revealed a large variance in each of the company performance variables of interest across all 4 years. It was evident from this initial data analysis that a few companies in the top 2.5 percentile (the conglomerates) were dominating the aggregated data, hence driving some of the measured trends. Notably, when we separated these companies from the dataset, the majority of the trends changed to provide a more consistent picture of steady growth.

The aim of the CSAT analysis was to quantify the size and growth of companies in the technology sectors to provide insight into their performance. DSIT wanted to understand companies that sat completely within the sector (SMEs) and isolate findings from larger companies. With this is in mind, for most of the analysis, DSIT and HMRC jointly developed an approach to isolate the large companies from the company performance dataset by categorising them based on 3 variables as described in the following section. This agreed approach isolated a similar sample of companies as the top 2.5 percentile.

To categorise a company as large we used the variables and criteria from the Companies Act SME definition. However, we adjusted this such that a company is classified if any one of the variables is categorised as large, as opposed to 2 of the 3. A company is therefore classified as large if any of the following is true in the first year that variable is reported:

• Employee count > 250
• Turnover > £36 million
• Tangible assets > £18 million

By requiring only one of these statements to be true, this approach is more likely to classify a company as large than the Companies Act would, ensuring that the conglomerate companies were captured as large. Notably, this also mitigates issues surrounding missing data, as this more liberal approach allowed more companies to be classified. Note that this is based on the end-of-year employee count.

Table 4 and Table 5 present the companies in each sector, classified as large and SME. In the results section, a comparison of large and SME companies is presented for some of the variables.

Table 4: Number of Large Companies, 2018-2019 to 2022-2023

Year	Eng Bio	Quantum	RAS
2018-2019	54	10	36
2019-2020	54	10	37
2020-2021	55	10	37
2021-2022	55	10	37
2022-2023	56	10	37

Table 5: Number of SMEs, 2018-2019 to 2022-2023

Year	Eng Bio	Quantum	RAS
2018-2019	1,101	50	1,837
2019-2020	1,183	59	1,981
2020-2021	1,314	65	2,156
2021-2022	1,365	68	2,244
2022-2023	1,448	76	2,355

By contrast, sector estimates from the R&D dataset retained large companies in the results, but a breakdown by company size is provided.

4.2.5 Subgroups

For each company performance variable of interest, we investigated the performance of 2 different groups of companies.

1. All Reporting Companies. This group is all companies that are reporting that variable of interest. The number of companies reporting each sum will increase over the series as new companies become active. There are a varying number of companies reporting each variable.
   
   
2. Cohort. This is the group of companies that reported the variable in the year 2018-2019. The number of companies in this group may reduce over the series if companies cease to exist. We investigate this group of companies to distinguish between growth in the variable of interest and growth in the number of companies reporting that variable.

5. Results

The section below shows estimates for the variables covered. There are some uncertainties in the way these estimates were produced. For example, the sector identification approach is new, and the process involved annualising reported data, matching information across different systems to get employee counts, making assumptions about company size, and relying on administrative data that may be revised later. Because of these factors, the estimates should be interpreted with caution. Whilst these measures are intended to be more reliable than alternatives, we are clear that the methods and data sources mean these measures are approximations.

5.1 Engineering Biology (Eng Bio)

The Eng Bio sector is split into supply chain and application sub-sectors, given the nature of companies. This is further explained in the Annex.

5.1.1 Company Performance Estimates

5.1.1.1 Turnover

Figure 1: Total Annual Turnover, UK Eng Bio (all companies), accounting periods ending in 2018-2019 to 2022-2023

Figure 1 shows that there was growth in the estimated total annual turnover for both large and SME companies. Turnover for large companies was estimated to be £34 billion in 2022-2023 whilst the turnover for SME companies was £2.6 billion.

Figure 2: Total Annual Turnover, UK Eng Bio SMEs (all reporting companies), accounting periods ending in 2018-2019 to 2022-2023

Figure 2 shows that, for SMEs, the total annual turnover in the sector for company accounting periods ending in 2022-2023 was estimated to be £1.7 billion for the supply chain companies, £570 million for the application companies, and £301 million for those in both groups. In total, this sums to £2.6 billion for Eng Bio SMEs.

Figure 3: Total Annual Turnover, UK Eng Bio SMEs (2018-2019 Cohort), accounting periods ending in 2018-2019 to 2022-2023

Figure 3 shows the estimated turnover over time for companies that reported on accounting periods ending in 2018-2019. Throughout the 2018-2019 to 2022-2023 period, the companies in this 2018-2019 cohort increased their Annual Turnover by an average of 24% per year for application companies, 7% per year for supply-chain companies, and 25% per year for those in both groups.

5.1.1.2 Employment

Figure 4 shows the employee count for all UK Eng Bio companies broken down by whether they have been classified as large or SME.

Figure 4: Employee Count, UK Eng Bio (all companies) in tax years 2018-2019 to 2022-2023

The estimated number of employees increased for Eng Bio SMEs across the period from 11,020 in the 2018-2019 PAYE tax year to 17,790 in 2022-2023. However, the estimated total number of employees for large companies fell from 36,320 in 2020-2021 to 30,550 in 2021-2022 before increasing to 32,700 in 2022-2023.

Figure 5: Employee Count, UK Eng Bio SMEs (all reporting companies) in tax years 2018-2019 to 2022-2023

Figure 5 shows that for SMEs, the estimated Employee Count in the Eng Bio sector increased for all company categories in the time-series.

The employee count in the Eng Bio sector in the tax year 2022-2023 was estimated to be 17,790 for SMEs, of which 7,790 were employed by the application companies, 8,230 by the supply chain companies, and 1,770 by those in both groups.

Figure 6: Employee Count, UK Eng Bio SMEs (2018-2019 Cohort), in tax years 2018-2019 to 2022-2023

Throughout the period, the companies in the 2018-2019 cohort were estimated to increase their employee count by an average of 9% per year (on a PAYE tax year basis) for application companies, 3% for supply-chain companies, and 14% for those in both groups.

5.1.1.3 Median Employment

Figure 7: Median Employee Count, UK Eng Bio (all companies), in tax years 2018-2019 to 2022-2023

Figure 7 shows that estimates of median employment stayed broadly flat for large companies at around 250 employees per company, whilst SME companies’ median employment grew slightly from 6 in the 2018-2019 PAYE tax year to 8 in 2022-2023.

Figure 8: Median Employee Count, UK Eng Bio SMEs (all reporting companies), in tax years 2018-2019 to 2022-2023

Figure 8 shows that for SMEs, median employee count was estimated to stay relatively consistent over the time series for all reporting companies.

The median employee count in the sector in the 2022-2023 PAYE tax year was estimated to be 8 for the application companies, 7 for the supply-chain companies, and 13 for those in both groups.

Figure 9: Median Employee Count, UK Eng Bio SMEs (2018-2019 Cohort), in tax years 2018-2019 to 2022-2023

From 2018-2019 to 2022-2023 (on a PAYE tax year basis) the companies in the 2018-2019 cohort increased their median employee count by a total of 5 for application companies, 1 for supply-chain companies, and 10 for those in both groups.

5.1.1.4 Total Expenditure on Wages

Figure 10: Total Expenditure on Wages, UK Eng Bio (all companies), in tax years 2018-2019 to 2022-2023

Figure 10 shows that wage expenditure was estimated to increase over the period for both large and SME companies. Total expenditure on wages for large companies in the 2022-2023 PAYE tax year was £2.7 billion.

Figure 11: Total Expenditure on Wages, UK Eng Bio SMEs (all reporting companies), in tax years 2018-2019 to 2022-2023

Figure 11 shows that for SMEs, total expenditure on wages in the sector was estimated to increase for all sub-groups.

Throughout the period, the total expenditure on wages in the sector increased by an average of 23% per year for application companies, 13% per year for supply-chain companies, and 30% per year for those in both groups.

The total expenditure on wages in the sector in the 2022-2023 PAYE tax year was estimated to be around £1.1 billion for Eng Bio SMEs; £399 million for the supply chain companies, £557 million for the application companies, and £112 million for those in both groups.

5.1.1.5 Average Employee Pay

The section below shows estimates for the variables covered. There are some uncertainties in the way these estimates were produced. For example, the sector identification approach is new, and the process involved annualising reported data, matching information across different systems to get employee counts, making assumptions about company size, and relying on administrative data that may be revised later. Because of these factors, the estimates should be interpreted with caution.

Table 6: Average Employee Pay (£), UK Eng Bio Large companies, in tax years 2018-2019 to 2022-2023

PAYE tax year	Number of Companies	Average Pay (£)
2018-2019	43	55,000
2019-2020	43	55,000
2020-2021	46	63,000
2021-2022	45	71,000
2022-2023	46	72,000

Table 7: Average Employee Pay (£), UK Eng Bio SMEs, in tax years 2018-2019 to 2022-2023

PAYE tax year	Number of companies	Average Pay (£)
2018-2019	644	40,000
2019-2020	713	43,000
2020-2021	744	46,000
2021-2022	829	46,000
2022-2023	826	48,000

Average employee pay findings for active companies is based on the pay of the total number of employees across the tax year. This cannot be directly compared with the employment count provided in section 5.1.1.2 which is based on an end of year employee count snapshot or the wages figures presented section 5.1.1.4 which include additional information as well as employee pay (see section 6.3 in the Annex for more information on all these variables). Though these numbers give you a rough sense of it being a significant sample size.

Tables 6 and 7 show that average employee pay was estimated to grow for both large and SME companies. Average employee pay was £72,000 for large companies and £48,000 for SME companies in the 2022-2023 PAYE tax year.

Table 8: Average Employee Pay (£), UK Eng Bio SMEs (all reporting companies), in tax years 2018-2019 to 2022-2023

Company type	2018-2019	2019-2020	2020-2021	2021-2022	2022-2023
Application	54,000	57,000	58,000	55,000	58,000
Supply chain	32,000	34,000	37,000	38,000	39,000
Both	40,000	41,000	43,000	43,000	51,000

As shown in table 8, for all reporting SMEs, average employee pay in the sector in the 2022-2023 PAYE tax year was estimated to be £58,000 for application companies, £39,000 for supply-chain companies, and £51,000 for those in both groups.

Between tax years 2018-2019 and 2022-2023, average employee pay in the sector was estimated to increase by £4,000 for application companies, £7,000 for supply-chain companies, and £11,000 for those in both groups.

Table 9: Average Employee Pay (£), UK Eng Bio companies (2018-2019 Cohort), in tax years 2018-2019 to 2022-2023

Company type	2018-2019	2019-2020	2020-2021	2021-2022	2022-2023
Application	54,000	58,000	61,000	59,000	59,000
Supply chain	32,000	34,000	37,000	37,000	39,000
Both	40,000	40,000	42,000	45,000	50,000

As shown in table 9, for the 2018-2019 cohort, average employee pay in the sector in the 2022-2023 PAYE tax year was estimated to be £59,000 for application companies, £39,000 for supply-chain companies, and £50,000 for those in both groups.

Between tax years 2018-2019 and 2022-2023, average employee pay was estimated to increase by £5,000 for application companies, £7,000 for supply chain companies, and £10,000 for those in both groups.

5.1.1.6 Partially Gross Trading Profit (PGTP)

Partially gross trading profit presents a particularly interesting story for these nascent sectors where, due to the nature of companies covered, several companies are loss-making. This is explained further in section 6.7 in the Annex.

Figure 12: Partially Gross Trading Profit, UK Eng Bio (all companies), accounting periods ending in 2018-2019 to 2022-2023

As shown in figure 12, whilst large Eng Bio companies were profit-making, SMEs were loss-making. For company accounting periods ending in 2022-2023, it was estimated that large companies had £2.2 billion in PGTP, whilst SMEs had -£3.3 billion.

Figure 13: Partially Gross Trading Profit, UK Eng Bio SMEs (all reporting companies), accounting periods ending in 2018-2019 to 2022-2023

As shown in figure 13, all reporting SMEs in the supply-chain were estimated to be profitable and increased their total PGTP by an average of 40% per year from 2018-2019 to 2021-2022, although there was a decrease for accounting periods ending in 2022-2023. The application companies and those in both groups were estimated to be in loss-makers and increased these losses by an average of 30% and 29% per year, respectively. This result remains similar for the 2018-2019 cohort.

The total PGTP in the sector for accounting periods ending in the most recent available financial year (2022-2023) was estimated to be £62 million for the supply-chain companies, -£3,142 million for the application companies, and -£261 million for those in both groups.

Given that the company list is a snapshot in time, increasing losses in later years may be a result of fewer companies reporting data in earlier years, or companies may actually be increasing their losses, or both. It is not known which of these factors is driving the results, however table 10 shows the increase in the number of companies reporting net losses in more recent years but also shows the increase in the number of companies reporting net profits.

Table 10: Number of UK Eng Bio SMEs Reporting Net Profits/Losses, accounting periods 2018-2019 to 2022-2023

Accounting Period	No. of SMEs reporting Net Losses	No. of SMEs reporting Net Profits
2018-2019	466	420
2019-2020	526	453
2020-2021	540	485
2021-2022	639	530
2022-2023	681	550

To further understand how Eng Bio companies survive whilst making losses, we show the breakdown of PGTP by whether companies are equity funded or not. This separates the companies into those who have received equity funding previously and those who have not. Figures 14 & 15 show the breakdown of PGPT by whether the companies are equity funded or not.

Figure 14: Partially Gross Trading Profit, UK Eng Bio SMEs (not equity funded), accounting periods ending in 2018-2019 to 2022-2023

Figure 15: Partially Gross Trading Profit, UK Eng Bio SMEs (equity funded), accounting periods ending in 2018-2019 to 2022-2023

As shown in figures 14 and 15, the group of Eng Bio SMEs that have received funding were estimated to be making larger losses than those who were not equity funded whose losses were increasing at a slower rate. As shown by table 11, it’s worth noting that there were less companies who are equity funded, although their overall losses were larger than companies who are not equity funded.

Table 11: Number of UK Eng Bio companies by equity funding status, accounting period ending in 2022-2023

Sub-Sector	Equity Funded	Not Equity Funded
Application	227	358
Supply Chain	51	516
Both	21	58
Total	293	932

5.1.1.7 Gross Value Added (GVA)

The story on Eng Bio companies’ GVA mirrored the PGTP data. Please note that for the purposes of this analysis, GVA is estimated using both CT and PAYE variables, which have different timing bases, so while we refer to financial years, the economic activity underpinning the reported tax data may not necessarily have taken place within the financial year (see section 6.5 in the Annex for further detail on data timings).

Figure 16: GVA, UK Eng Bio (all companies), 2018-2019 to 2022-2023

Figure 16 shows that, like PGTP, large companies were estimated to have positive GVA, whereas SMEs were estimated to have negative GVA. Large companies had £5 billion GVA in 2022-2023, whereas SMEs had -£1.8 billion.

Figure 17: GVA, UK Eng Bio SMEs (all reporting companies),2018-2019 to 2022-2023

As shown in figure 17, the total GVA in the sector in the most recent available financial year (2022-2023) was estimated to be £457 million for the supply-chain companies, -£2.1 billion for the application companies, and -£99 million for those in both groups. All reporting SMEs in the supply-chain were estimated to be in net-positive GVA and were increasing their total GVA by an average of 14% per year, although there was a decrease in 2022-2023 compared to 2021-2022. The application companies and those in both groups were estimated to be in net-negative GVA and were decreasing total GVA by an average of 29% and 50% per year, respectively.

Figures 18 and 19 show GVA for SME companies split between those who had received equity funding and those who had not.

Because equity funded companies were less likely to be profit-making, these companies had larger negative GVA than companies that had not previously received funding.

Figure 18: GVA, UK Eng Bio SMEs (not equity funded), 2018-2019 to 2022-2023

Figure 19: GVA, UK Eng Bio SMEs (equity funded), 2018-2019 to 2022-2023

Comparing figures 18 and 19, application companies that were not equity funded were estimated to make smaller negative GVAs than those that were equity funded.

The negative GVA contributions from applications companies that were equity funded, as shown in figure 19, were estimated to increase at twice the rate of companies who were not in receipt of fundraising, as shown in figure 18.

Figure 18 also shows that the companies that were in both groups that were not equity funded were estimated to increase their GVA, which turned from negative to positive GVA in 2020-2021. However, the equity funded companies in this same group had negative GVA throughout the period, which was increasing between years, as shown in figure 19.

5.1.2 R&D tax credits

5.1.2.1 Number of claimants

Figure 20: Number of UK Eng Bio companies that claimed R&D tax credits, and non-claimant companies for accounting periods ending in 2018-2019 to 2021-2022

Figure 20 shows that the number of Eng Bio companies claiming R&D tax credits (under either the SME or RDEC scheme) was estimated to increase steadily over the 4 years.

For accounting periods ending in the most recent financial year available (2021-2022), 538 Eng Bio companies were found to be claiming R&D (38% of the 1,420 active Eng Bio companies).

However, when grouping claims at URL level – since several CRNs may be grouped for different websites – the proportion of URLs in the company lists claiming R&D tax credits is higher, suggesting some companies may be claiming R&D tax credits on behalf of their group.

Figure 21: Number of UK Eng Bio URLs that claimed R&D tax credits, and non-claimant URLs for accounting periods ending in 2018-2019 to 2021-2022

Figure 21 shows that 50% of companies grouped at URL level claimed R&D tax credits in 2021-2022.

5.1.2.2 Qualifying expenditure by size

Figure 22: Total UK Eng Bio R&D qualifying expenditure by company size (large/SME) for accounting periods ending in 2018-2019 to 2021-22 (£ million)

Figure 22 shows that large companies accounted for most of the qualifying expenditure per year; however, expenditure by SME companies increased over the 4-year period, gradually evening out the balance.

For accounting periods ending in 2021-2022, it was estimated that SMEs had £980 million of qualifying expenditure, whilst large companies had £1,170 million.

5.1.2.3 Qualifying expenditure by subgroup

Figure 23: Total UK Eng Bio R&D qualifying expenditure by subgroup for accounting periods ending in 2018-2019 to 2021-2022 (£ million)

Application companies were estimated to invest the largest amount of qualifying expenditure to R&D investment, contributing 70% of the total Eng Bio expenditure in accounting periods ending in 2021-2022.

Qualifying expenditure for application companies saw a decrease in 2019-2020 to £1.3 billion, however recovered in 2020-2021 and 2021-2022 to £1.6 billion.

R&D investment was estimated to increase across the time series for supply chain companies, as well as companies classified as both.

5.1.2.4 Additional government support

Figure 24 shows the various combinations of direct government support received by application companies. This includes Innovate UK (iUK) grants, British Business Bank (BBB) backed equity programme funding or R&D tax credits. Those who did not receive funding from any of these sources may have received funding from other sources.

For all types of financial support except R&D, only one snapshot was available for the company being in receipt of support and the timing of the funding is unknown. For R&D, the figure reflects R&D reliefs pertaining to accounting periods ending in 2021-2022, which may have occurred before or after the funding from other government sources.

Figure 24: Number of UK Eng Bio companies that claimed R&D tax credits for accounting periods ending in 2021-2022, ever received BBB backed equity funding or ever received an iUK grant

The chart indicates that whilst a lot of Eng Bio companies did not appear to receive any government support, the most common combination of funding sources was an iUK grant paired with R&D tax credits, followed by R&D only, which indicates that R&D tax relief was one of the most popular sources of financial support in the Eng Bio sector.

5.2 Quantum

5.2.1 Company Performance Estimates

5.2.1.1 Turnover

Figure 25: Total Annual Turnover, UK Quantum (all companies), accounting periods ending in financial years 2018-2019 to 2022-2023

Figure 25 shows that large companies were much larger than SMEs in the Quantum sector. For accounting periods ending in 2022-2023, total turnover was estimated to be £22 billion for large Quantum companies compared to £216 million for SMEs.

Figure 26: Total Annual Turnover, UK Quantum SMEs (all reporting companies), accounting periods ending in 2018-2019 to 2022-2023

For SMEs, the total annual turnover in the Quantum sector grew across the series from £100 million in 2018-2019 to £216 million in 2022-2023.

Figure 27: Total Annual Turnover, UK Quantum SMEs (2018-2019 Cohort), accounting periods ending in 2018-2019 to 2022-2023

Figure 27 shows total annual turnover in the sector was estimated to increase by an average of 22% for the 2018-2019 cohort per year across the time-series.

5.2.1.2 Employment

Quantum companies’ employee data cannot be shown for all years due to statistical disclosure risks. Figure 28 shows the employee count for all UK Quantum companies broken down by whether they have been classified as large or SME.

Figure 28: Employee Count, UK Quantum (all companies), in tax years 2018-2019 to 2022-2023

Figure 28 shows the scale of larger companies compared to SMEs, with over 34,000 employees estimated for large companies versus 1,680 employees for SMEs in the 2022-2023 PAYE tax year. This large difference can be explained by the presence of large, diversified companies in the quantum company list; these are excluded when looking at SMEs only.

Figure 29: Employee Count, UK Quantum SMEs (all reporting companies), in tax years 2018-2019 to 2022-2023

As shown by figure 29, for all reporting SMEs, employee count in the sector was estimated to increase by an average of 20% each year across the time-series.

Figure 30: Employee Count, UK Quantum SMEs (2018-2019 Cohort), in tax years 2018-2019 to 2022-2023

It was estimated that the SME companies in the 2018-2019 cohort increased their employee count by an average of 15% per year from 820 in the 2018-2019 PAYE tax year to 1,430 in 2022-2023.

5.2.1.3 Median Employment

Median employment was also much higher for large companies than SMEs, however, data is not presented for large companies for PAYE tax years 2018-2019 to 2020-2021, because there were too few companies for this to be disclosed.

Figure 31: Median Employee Count, UK Quantum (all companies), in tax years 2018-2019 to 2022-2023

The median employee count in the 2022-2023 PAYE tax year was estimated to be 2,678 for large companies and 13 for SMEs.

Figure 32: Median Employee Count UK Quantum SMEs (all reporting companies), in tax years 2018-2019 to 2022-2023

For all reporting SMEs, the median employee count was estimated to increase from 6 in the 2018-2019 PAYE tax year to 13 in 2022-2023.

Figure 33: Median Employee Count, UK Quantum SMEs (2018-2019 Cohort), in tax years 2018-2019 to 2022-2023

As shown in figure 33, from the 2018-2019 PAYE tax year to 2022-2023, it was estimated that the SME companies in the 2018-2019 cohort increased their median employee count by a total of 14, from 6 to 20.

5.2.1.4 Total Expenditure on Wages

Figure 34: Total Expenditure on Wages, UK Quantum (all companies), in tax years 2018-2019 to 2022-2023

Despite numbers being too small to disclose for large companies until the 2021-2022 PAYE tax year, figure 34 shows that wages increased over the period for both large and SME companies. In the 2022-2023 tax year, total expenditure on wages for large companies was estimated to be £2.1 billion and £102 million for SMEs.

Figure 35: Total Expenditure on Wages, UK Quantum SMEs (all reporting companies), in tax years 2018-2019 to 2022-2023

For SMEs, it was estimated that total expenditure on wages in the sector increased by an average of 28% each year.

5.2.1.5 Average Employee Pay

Due to the small number of companies, data was not disclosed for average employee pay of large companies until the 2021-2022 PAYE tax year.

Table 12: Average Employee Pay (£), UK Quantum Large companies, in tax years 2021-2022 to 2022-2023

PAYE tax year	Number of Companies	Average Pay (£)
2021-2022	10	49,000
2022-2023	10	52,000

Table 13: Average Employee Pay (£), UK Quantum SMEs, in tax years 2018-2019 to 2022-2023

PAYE tax year	Number of Companies	Average Pay (£)
2018-2019	33	38,000
2019-2020	44	38,000
2020-2021	46	40,000
2021-2022	53	47,000
2022-2023	54	51,000

Average employee pay findings for active companies is based on the pay of the total number of employees across the tax year. This cannot be directly compared with the employment count provided in section 5.2.1.2 which is based on an end of year employee count snapshot or the wages figures presented section 5.2.1.4 which include additional information as well as employee pay (see section 6.3 in the annex for more information on all these variables).

As shown in tables 12 and 13, average employee pay grew for both large and SME companies and was estimated at a similar level of just above £50,000 for both large and SME companies in the 2022-2023 PAYE tax year. For all reporting SMEs, average employee pay was estimated to increase by £13,000 over the time series.

5.2.1.6 Partially Gross Trading Profit (PGTP)

There was no recorded PGTP data for large Quantum companies because there were too few companies to report.

Figure 36: Partially Gross Trading Profit, UK Quantum SMEs (all reporting companies), accounting periods ending in 2018-2019 to 2022-2023

As shown in figure 36, SMEs in the sector were estimated to be in loss-making and the total PGTP for SMEs in the sector in 2022-2023 was estimated to be -£91 million. Increasing losses in later years may be a result of the increasing number of companies with reported data in more recent years, or companies may be increasing their losses. It is not certain which of these factors is stronger.

5.2.1.7 Gross Value Added (GVA)

GVA was not available for large companies because there were too few companies to report.

Please note that for the purposes of this analysis, GVA is estimated using both CT and PAYE variables, which have different timing bases, so while we refer to financial years, the economic activity underpinning the reported tax data may not necessarily have taken place within the financial year (see the Annex for further detail on data timings).

Figure 37: GVA, UK Quantum SMEs (all reporting companies), accounting periods ending in 2018-2019 to 2022-2023

As shown in figure 37, for SMEs, total GVA was estimated to decrease from a net-positive to a net-negative between 2019-2020 and 2020-2021 and increased consistently per year since, reaching £12 million in 2022-2023.

5.2.2 R&D tax credits

5.2.2.1 Number of claimants

Figure 38 shows that the number of Quantum companies estimated to be claiming R&D tax credits (under either the SME or RDEC scheme) increased steadily over the 4 years.

Figure 38: Number of UK Quantum companies that claimed R&D tax credits, and non-claimant companies for accounting periods ending in 2018-2019 to 2021-2022

For accounting periods ending in the most recent financial year available (2021-2022), 53 Quantum companies were found to be claiming R&D (68% of the 78 active Quantum companies).

However, when grouping claims at URL level – since several CRNs may be grouped for different websites – the proportion of URLs in the company lists claiming R&D tax credits is higher.

Figure 39: Number of UK Quantum URLs that claimed R&D tax credits, and non-claimant URLs for accounting periods ending in 2018-2019 to 2021-2022

Figure 39 shows that over 70% of Quantum companies grouped at URL level claimed R&D tax credits in 2021-2022.

5.2.2.1 Additional government support

Figure 40 shows the various combinations of direct government support received by Quantum companies. This includes Innovate UK (iUK) grants, British Business Bank (BBB) backed equity programme funding or R&D tax credits. Those who did not receive funding from any of these sources may have received funding from other sources.

For all types of financial support except R&D, only one snapshot was available for the company being in receipt of support and the timing of the funding is unknown. For R&D, figure 40 reflects R&D reliefs pertaining to accounting periods ending in 2021-2022, which may have occurred before or after the funding from other government sources.

Figure 40: Number of UK Quantum companies that claimed R&D tax credits in accounting periods ending in 2021-2022, ever received BBB backed equity funding or ever received an iUK grant

The most common combination of funding sources was an iUK grant paired with R&D tax credits, followed by R&D only, which indicates that R&D is one of the most popular sources of financial support in the Quantum sector. 28 companies received both an iUK grant and R&D tax relief for accounting periods ending in 2021-2022.

5.3 Robotics and Autonomous Systems (RAS)

Companies operating in the manufacture, development or integration of robotics and autonomous systems (RAS) are included in this analysis. This is further detailed in section 6.6 in the Annex.

5.3.1 Company performance estimates

5.3.1.1 Turnover

Figure 41: Total Annual Turnover, UK RAS (all companies), accounting periods ending in 2018-2019 to 2022-2023

Figure 41 shows that there was growth in turnover for large companies until 2022-2023 where there was a small decline compared to the previous year. For accounting periods ending in 2022-2023, total turnover was estimated to be £9.2 billion for large RAS companies.

Figure 42: Total Annual Turnover, RAS SMEs (all reporting companies), accounting periods ending in 2018-2019 to 2022-2023

The total annual turnover in the sector for accounting periods ending in 2022-2023 was estimated to be £3.9 billion for RAS SMEs, and annual turnover increased by an average of 11% per year over the period.

Figure 43: Total Annual Turnover, UK RAS SMEs (2018-2019 Cohort), accounting periods ending in 2018-2019 to 2022-2023

It was estimated that the companies in the 2018-2019 cohort increased their annual turnover by an average of 8% per year from £2.6 billion in 2018-2019 to £3.5 billion in 2022-2023.

5.3.1.2 Employment

Figure 44: Employee Count, UK RAS (all companies), in tax years 2018-2019 to 2022-2023

Figure 44 shows the number of employees increased for both large and SME RAS companies across the period. There were 47,970 estimated employees reported for large companies in the 2022-2023 PAYE tax year.

Figure 45: Employee Count, UK RAS SMEs (all reporting companies), in tax years 2018-2019 to 2022-2023

For all reporting SMEs, there were 23,420 estimated employees in the 2022-2023 PAYE tax year.

Figure 46: Employee Count, UK RAS SMEs (2018-2019 Cohort), in tax years 2018-2019 to 2022-2023

It was estimated that the companies in the 2018-2019 cohort increased their employee count by an average of 5% per year.

5.3.1.3 Median employment

Figure 47: Median Employee Count, UK RAS (all companies), in tax years 2018-2019 to 2022-2023

For large companies, it was estimated that median employment peaked at over 400 employees in the 2021-2022 PAYE tax year, before falling to 394 in 2022-2023.

Figure 48: Median Employee Count, UK RAS SMEs (all reporting companies), in tax years 2018-2019 to 2022-2023

Figure 48 shows that for RAS SMEs, the median employee count estimates stayed consistent at 5 until the 2022-2023 PAYE tax year when this grew to 6.

5.3.1.4 Total Expenditure on Wages

Figure 49: Total Expenditure on Wages, UK RAS (all companies), in tax years 2018-2019 to 2022-2023

Apart from a small decline in the 2020-2021 PAYE tax year for large RAS company wages, wage estimates increased over the period for both large and SME companies. For large companies, total wages were estimated to be £2.1 billion in the 2022-2023 PAYE tax year.

Figure 50: Total Expenditure on Wages, UK RAS SMEs (all reporting companies), in tax years 2018-2019 to 2022-2023

For SMEs, total expenditure on wages in the sector increased in the period. Total wage expenditure in the 2022-2023 PAYE tax year for SMEs was estimated to be £1.1 billion, as shown in figure 51.

For all reporting SMEs, the estimated total expenditure on wages in the sector increased by an average of 15% per year.

5.3.1.5 Average Employee Pay

Table 14: Average Employee Pay (£), UK RAS Large companies, in tax years 2018-2019 to 2022-2023

PAYE tax year	Number of Companies	Average Pay (£)
2018-2019	28	30,000
2019-2020	30	32,000
2020-2021	30	30,000
2021-2022	30	31,000
2022-2023	30	33,000

Table 15: Average Employee Pay (£), UK RAS SMEs, in tax years 2018-2019 to 2022-2023

PAYE tax year	Number of Companies	Average Pay (£)
2018-2019	1,130	30,000
2019-2020	1,207	32,000
2020-2021	1,261	33,000
2021-2022	1,378	37,000
2022-2023	1,348	37,000

Average employee pay findings for active companies is based on the pay of the total number of employees across the tax year. This cannot be directly compared with the employment count provided in section 5.3.1.2 which is based on an end of year employee count snapshot or the wages figures presented section 5.3.1.4 which include additional information as well as employee pay (see section 6.3 in the Annex for more information on all these variables).

Tables 14 and 15 show that average employee pay estimates were highest in 2022-2023 for both large companies and SMEs, although there was dip in large companies’ average employee pay in 2020-2021 and 2021-2022.

In the 2022-2023 tax year, average employee pay estimates were £33,000 for large companies and £37,000 for SMEs. SME average employee pay was also higher in the 2020-2021 and 2021-2022 PAYE tax years.

For all reporting SMEs, average employee pay in the sector was estimated to increase by £7,000 over the period.

5.3.1.6 Partially Gross Trading Profit (PGTP)

Whilst large companies were profit-making, SMEs were loss-making.

Figure 51: Partially Gross Trading Profit, UK RAS (all companies), accounting periods ending in 2018-2019 to 2022-2023

Figure 51 shows that large companies were profit-making from 2019-20 onwards; however, levels have varied over time. For accounting periods ending in 2022-2023, large companies’ PGTP was estimated to be £159 million.

Figure 52: Partially Gross Trading Profit for UK RAS SMEs, accounting periods ending in 2018-2019 to 2022-2023

Figure 52 shows, for all reporting SMEs, losses were estimated to be £381 million for accounting periods ending in 2022-2023, and profits fell by an average of 70% through the period.

Given that the company list is a snapshot in time, increasing losses in later years may be a result of fewer companies reporting data in earlier years, or companies’ losses may be increasing, or both. It is not known which of these factors is driving the results. However, table 16 shows the increase in the number of companies reporting net losses in more recent years but also shows the increase in the number of companies reporting net profits. There was a smaller number of companies reporting net losses, but these losses were larger than profits made by profit-making companies, to provide the results in figure 53.

Table 16: Number of UK RAS SMEs Reporting Net Profits/Losses, accounting periods 2018-2019 to 2022-2023

Accounting Period	No. of SMEs reporting Net Losses	No. of SMEs reporting Net Profits
2018-2019	493	1,037
2019-2020	585	1,111
2020-2021	662	1,156
2021-2022	754	1,186
2022-2023	744	1,296

Figures 53 and 54 show PGTP results for whether RAS SMEs were equity funded or not.

Figure 53: Partially Gross Trading Profit, UK RAS SMEs (not equity funded), accounting periods ending in 2018-2019 to 2022-2023

The group that is not equity funded were estimated to be profit-making for all periods in the time series apart from accounting periods ending in 2021-2022. Meanwhile, as shown in figure 54, the group that was equity funded were driving the increase in losses.

Figure 54: Partially Gross Trading Profit, UK RAS SMEs (equity funded), accounting periods ending in 2018-2019 to 2022-2023

Although a smaller proportion of RAS companies were equity funded, these companies contributed the largest losses to the overall losses of the sector. Table 17 shows the sample sizes for companies receiving equity funding or not in 2022-2023, suggesting only 10% of companies were equity funded but the losses of these companies were large enough to produce the result in figure 52.

Table 17: Number of UK RAS companies – by equity funding status, 2022-2023

Total RAS Companies	Equity Funded	Not Equity Funded
2,040	200	1,840

5.3.1.7 Gross Value Added (GVA)

GVA contributions were positive across the time series for both large RAS companies and SMEs. This was mostly because the total expenditure on wages for RAS companies outweighed the relatively small losses for SMEs. Together, GVA was estimated to total to over £2.6 billion in 2022-2023.

Figure 55: GVA, UK RAS (all companies), 2018-2019 to 2022-2023

Figure 55 displays the growth in GVA across the period. For large companies, GVA was estimated to be £1.9 billion for companies reporting in 2022-2023.

Please note that for the purposes of this analysis, GVA is estimated using both CT and PAYE variables, which have different timing bases, so while we refer to financial years, the economic activity underpinning the reported tax data may not necessarily have taken place within the financial year (see the Annex for further detail on data timings).

Figure 56: GVA, UK RAS SMEs (all reporting companies), 2018-19 to 2022-2023

Figure 56 shows for all reporting SMEs, the sector was in net-positive GVA and increased by an average of 9% annually. Total GVA in the 2022-2023 period was estimated to be £759 million for SMEs.

Figures 57 and 58 breakdown the GVA of RAS SMEs by whether the companies received equity funding or not.

Figure 57: GVA, UK RAS SMEs (not equity funded), 2018-2019 to 2022-2023

Shown in Figure 57, the companies that were not equity funded were estimated to be in a net-positive GVA and this increased by an average of 14% per year. By comparison, GVA for the companies that were equity funded was estimated to be negative, decreasing by 16% on average per year to -£118 million in 2022-2023.

Figure 58: GVA, UK RAS SMEs (equity funded), 2018-2019 to 2022-2023

This result reflects the profit results, where the losses are driven by a smaller sample of RAS companies, however, companies that are not equity funded had positive GVA, likely due to higher wages.

5.3.2 R&D tax credits

5.3.2.1 Number of claimants

Figure 59: Number of UK RAS companies that claimed R&D tax credits, and non-claimant companies for accounting periods ending in 2018-2019 to 2021-2022

The number of RAS companies claiming R&D tax credits (under either the SME or RDEC scheme) was estimated to increase steadily over the 4 years, albeit not as quickly as the number of active companies in the sector overall.

For accounting periods ending in the most recent financial year available (2021-2022), 598 RAS companies were estimated to be claiming R&D (26% of the 2,281 active RAS companies).

However, when grouping claims at URL level – since several CRNs may be grouped for different websites – the proportion of URLs in the company lists claiming R&D tax credits is higher.

Figure 60: Number of UK RAS URLs that claimed R&D tax credits, and non-claimant URLs for accounting periods ending in 2018-2019 to 2021-2022

Figure 60 shows that over 35% of companies grouped at URL level claimed R&D tax credits in 2021-2022.

5.3.2.2 Qualifying expenditure

Figure 61: Total UK RAS R&D qualifying expenditure by company size (large/SME) for accounting periods ending in 2018-2019 to 2021-2022 (£ million)

Large companies accounted for most qualifying expenditure per year; however, expenditure by SME companies increased over the 4-year period.

During accounting periods ending in the 2021-2022 financial year, it was estimated that SMEs had approximately £350 million of qualifying expenditure, whilst large companies had approximately £530 million.

5.3.2.3 Additional government support

Figure 62 shows the various combinations of direct government support received by RAS companies. This includes Innovate UK (iUK) grants, British Business Bank (BBB) backed equity funding or R&D tax credits. Those who did not receive funding from any of these sources may have received funding from other sources.

For all types of financial support except R&D, only one snapshot was available for the company being in receipt of support and the timing of the funding is unknown. For R&D, the figure reflects R&D reliefs pertaining to accounting periods ending in 2021-2022, which may have occurred before or after the funding from other government sources.

Figure 62: Number of UK RAS companies that claimed R&D tax credits in accounting periods ending in 2021-2022, ever received BBB backed equity funding or ever received an iUK grant

Figure 62 indicates that whilst a lot of RAS companies did not appear to receive any government support, the most common funding source was R&D tax credits only, followed by a combination of iUK grant paired with R&D tax credits, which indicates that R&D tax relief is one of the most popular sources of financial support in the RAS sector.

6. Annex

6.1 HMRC Company Summary Across Taxes (CSAT) database

6.1.1 Overview

CSAT is a collection of datasets or tables about companies and the taxes they pay. The datasets can be combined to build a rich picture of the corporate population and the business environment in which they operate. CSAT contains data from multiple tax heads such as Pay As You Earn (PAYE), Corporation Tax (CT) and Value Added Tax (VAT).

CSAT includes all active entities registered for CT and those which have ceased since April 2015: 9.9 million companies in January 2024, of which 5.4 million are active.

The CSAT database does not include entity registration data for partnerships, sole proprietorships, or other entity types not required to register for CT (including non-profit making charities, public bodies, and trusts).

6.1.2 Matching across tax heads

HMRC customer data is stored in different, independently designed databases, without a single unique customer identifier (ID) recorded in all systems to allow record linking.

Links between tax heads are derived in multiple ways, one being a Standardised Name Matching (SNM) process, originally implemented in 2015, but incrementally improved since, to standardise the names of companies, corporate VAT traders and employers to link their tax registration records. SNM is an analytical data matching approach (as it does not use identifiers) but is stricter than fuzzy matching so sacrifices coverage for accuracy. For most companies where the registered name is sufficiently similar on different systems, the standardised name can be regarded as a master customer key that can be used to create lookup tables between the HMRC issued tax reference numbers captured on different systems.

Instead of relying solely on standardised name matching to provide most links between a company and the payroll registration, CSAT integrates and prefers other data sources that capture links provided directly by the customer from a variety of registration, submission, and operational processes. They are categorised by customer-provided or HMRC-derived links and links are cross referenced across multiple sources to validate them.

CSAT contains links between:

• CT Unique Tax Reference (UTR) and VAT Registration Number (VRN).
• CT UTR and payroll reference number for companies that employ staff.

The Companies House-issued Company Registration Number (CRN) is collected in the CT data, so no analytical matching is necessary between CT UTR and CRN.

The links created by SNM are supplemented by additional links created as by products of existing HMRC processes designed to administer income tax.

Many companies do not run a payroll scheme, but of those that do, the matching accuracy and coverage for company to payroll links is 98%. Of the company to payroll links established and accepted in the database, 98% of those are between one company and one payroll scheme.

6.1.3 Employee count variable

The Real Time Information (RTI) payroll submissions to HMRC, from which the employee count variables are derived, do not contain information on the number of hours worked by each employee, so will include all employees paid regardless of how many hours they have worked. Depending on the extent to which the company employs part-time staff, the total employee count from RTI may be higher than a full-time equivalent figure. RTI submissions will also show employments that have not lasted a full year.

6.1.4 Annualisation

Unlike Income Tax that follows the tax year (6 April to 5 April) and consequently payroll submission data in RTI, companies submitting Corporation Tax returns and traders submitting VAT returns are allowed to define their own accounting periods (APs) within certain conditions. For instance, CT APs can be of any length of time but must not exceed 12 months.

CSAT uses the end-date of the AP to allocate the values (turnover, profit etc.) to a financial year (FY, 1 April to 31 March). Where a company has multiple APs ending in a single FY – e.g. 1 January to 30 June 2017 and 1 July 2017 to March 2018 – both are allocated to the 2017-2018 FY despite covering a period of more than 12 months.

The second challenge of annualising CT return values is handled after the AP has been allocated to the FY. The 2 APs in the above example cover 450 days. To arrive at the annualised amount, we divide the aggregated CT return values from both returns by this number of days (to derive a daily figure) and then multiply this by 365 days to obtain the annualised figure.

6.1.5 Data availability and amendments

CT600 return data for the last 7 completed financial years is updated annually each May in CSAT. This assigns returns filed with an accounting period (AP) end date falling between 1 April and 31 March to that year. Companies can choose their own AP but must submit their CT return one year after the AP end date. For example, a company with an AP ending 31 December 2022 should file by 31 December 2023. As of July 2024, when the data used in this publication was extracted, the latest completed financial year was 2022-23, filed by 31 March 2024. At this point, the time lag will range between 1 and 2 years depending on the AP end date.

Following the filing deadline, a company may:

• File their return late, if they missed the original deadline
• File a revised return (for example, to correct an error)
• File an amended return (for example, following an enquiry, where HMRC has reviewed the return and determined that changes are necessary. This process can take several years. Or from companies with an updated declaration of their profits/liability for the year (without HMRC investigation)).

6.2 HMRC Companies Database (CDB)

The Companies Database (CDB) provides summary information on companies’ CT liabilities, for accounting periods (APs) ending in a given financial year. It underpins HMRC’s National Statistics on CT, but figures from the CDB presented in this report won’t be directly comparable due to variations in variable definitions and due to the specific sector populations used.

Unlike CSAT, the CDB covers mainly active companies; it is based on the data submitted via CT600 tax returns. It is more up to date than CSAT, in that it takes the latest available tax return information for each company, including capturing the company’s latest tax return amendment. It also augments the data with extra information from other data sources, including the ONS’ Interdepartmental Business Register (IDBR), Companies House, Bureau van Dijk FAME , and ONS geographical data, but does not link the CT data to other tax data.

Furthermore, unlike CSAT data, the CDB is not annualised, so where companies have more than one AP ending in a single financial year, this will appear as a single AP, which may be shorter or longer than 12 months in total, i.e. figures may differ from equivalent annualised figures in CSAT. CDB figures may also differ from CSAT figures due to the times at which each database is updated; all the data used in this report is administrative tax data and is subject to ongoing revisions and amendments.

The CDB was used in this analysis in addition to CSAT as CSAT does not contain information on Capital Allowances and Balancing Charges, so we combined data from CDB and CSAT to produce the measure of PGTP (see the next sections), although CSAT was prioritised as the main source of tax data.

6.2.1 Capital Allowances

Investment costs are not automatically factored into a business’s profit/loss calculations for Corporation Tax. Capital allowances (CAs) exist to incentivise investment and can therefore be used to reduce a business’s taxable profit. They provide relief on either the full value of the investment or at an assumed depreciation rate claimed over many years. The major types of qualifying investment are plant and machinery, including vehicles. Some other types of assets can be claimed on through specific types of CA, most notably the Structures and Buildings Allowance.

When a company makes a CA claim, a percentage of the qualifying expenditure can be claimed against profits in that tax year (the percentage depends on the type of CA being claimed). The amount of expenditure that remains after this claim is termed the pool and this is carried forward into the following years to be claimed on against future profits. Any new investments are added to the pool. If a company sells an asset that has previously been claimed on through CAs, the amount it sells for is the disposal value. This disposal value can result in a balancing charge, where the amount that arises from the sale is more than the balance of the pool and is therefore added to the company’s taxable profits.

6.3 Key Variables and Derived Statistics

HMRC extracted the data necessary to calculate the key variables of interest that we report in this publication. Note that the analysis does not account for inflation on cash variables.

Annual Turnover

Annual Turnover is the trading turnover earned during the accounting period ending in the financial year, reported in box 145 of the CT600 return, annualised according to that company’s accounting period for the given year (if the accounting period length is not a year).

Employee variables - this includes Employee Count, Expenditure on Wages, and Average Employee Pay. These are sourced from Pay As You Earn (PAYE) Real Time Information (RTI) data. These employees will be a mix of part-time and full-time workers and the balance between them will vary by company.

Employee Count

The variable used to measure employee count is the number of active National Insurance Numbers (NINos) that were paid and remain part of that scheme at the end of the tax year. This “end-of-year” employee count snapshot is presented for each sector, as well as the median employee count based on the “end-of-year” values. In a small number of cases, if the “end-of-year” employee count is not available this has been supplemented with the total employee count across the year.

Expenditure on Wages

Expenditure on wages is quantified as:

Wages = employee pay + pension contributions + other staff costs, (equation 1)

where ‘employee pay’ is a company’s total expenditure on staff wages and salaries, ‘pension contributions’ is employee pension contribution , and ‘other staff costs’ includes examples such as statutory sick or maternity pay, ordinary or additional paternity pay, and adoption pay. Note that we do not have an employer pension contribution variable in the datasets. Also, note that these variables reflect the total across the tax year rather than limited to those employees who remain part of the scheme at the end of the tax year.

Average Employee Pay

Average Employee pay for the sector is calculated as:

Average Employee Pay = sum total employee pay / sum total employee count, (equation 2)

where ‘sum total employee pay’ is the sum of all companies’ expenditure on employee pay in the sector across the tax year and ‘sum total employee count’ is the sum of all companies’ employee counts in the sector across the tax year. Note, these variables reflect the total across the tax year rather than limited to those employees who remain part of the scheme at the end of the tax year.

Partially Gross Trading Profit (PGTP)

We chose PGTP as the key indicator of profit as it has been specifically constructed to include capital allowance claims and balancing charges. Capital allowances (CAs) are a form of tax relief on capital expenditure with the purpose of incentivising investment , whilst Balancing Charges (BC’s) arise when an asset is sold for more than its recorded CA claim value (see Capital Allowances section. Net Trading Profit has these deducted from trading profit to reduce the company’s taxable profit. However, CAs are a positive indicator of the health and growth of investment-based companies like those in this report. We calculate PGTP as:

PGTP = Net Trading Profit + CAs - BCs (equation 3)

where the CAs are the sum of all the capital allowances a company has claimed, and the BCs are the sum of all the balancing charges a company has incurred under their CA claims. The Net Trading Profit is that of a company’s declared net trading profit in the accounting period of interest i.e. the total trading profits less the total trading losses from the same accounting period. Note that the company may have carried some of these trading losses forward or backward to increase/ decrease their taxable income and this is not accounted for under this definition of net trading profit.

Capital Allowance & Balancing Charges variables are taken from CDB rather than CSAT. To appropriately combine the capital allowance claims to calculate PGTP using equation 3 we first annualised the CAs and BCs as follows,

CAann = CAap / ap length × 365, (equation 4)

Gross Value Added

This variable is an approximation for GVA, estimating the value of a company’s outputs based on the limited relevant information collected via the tax system.

GVA = PGTP + Wages, (equation 5)

where Wages is defined above in equation 1 .

The tables below provide detailed statistics on the coverage of each variable

Table 18: Eng Bio company performance variables coverage, 2018-2019 to 2022-2023

Year	Annual Turnover	GVA	PGTP	Employee Variables
2018-2019	71%	57%	81%	60%
2019-2200	72%	59%	83%	61%
2020-2021	70%	57%	83%	58%
2021-2022	70%	60%	86%	62%
2022-2023	69%	57%	85%	58%

Table 19: Quantum company performance variables coverage, 2018-2019 to 2022-2023

Year	Annual Turnover	GVA	PGTP	Employee Variables
2018-2019	75%	68%	80%	70%
2019-2020	77%	74%	84%	77%
2020-2021	80%	73%	88%	73%
2021-2022	80%	77%	89%	81%
2022-2023	74%	73%	88%	74%

Table 20: RAS company performance variables coverage, 2018-2019 to 2022-2023

Year	Annual Turnover	GVA	PGTP	Employee Variables
2018-2019	81%	60%	84%	62%
2019-2020	83%	60%	86%	61%
2020-2021	82%	58%	84%	59%
2021-2022	83%	61%	87%	62%
2022-2023	83%	57%	87%	58%

The coverage of each variable varies year on year and across the sectors. This is expected, as an active company will not always report every variable of interest in the dataset. Likely reasons for the coverage differences in the variables of interest are:

1. Annual Turnover – this variable is drawn from Corporation Tax return data, but it is not a mandatory field on the CT600 form where it is collected.
   
2. GVA – this variable is calculated from PGTP and employee variables (see equation 5) and therefore has lower coverage than these 2 variables alone.
   
3. PGTP – this variable is also drawn from CT600 data. Notably it is calculated from trading profit and loss variables, as opposed to taxable profit and loss variables. As such, a company will only report the components of this variable if they are trading. A company which is doing research but is not trading may not report the components needed to calculate this variable .
   
4. Coverage for CT variables is technically around 85% across all 3 sectors as HMRC has received tax returns from these companies in 2022-23 (though this is lower in earlier years). In this analysis we have attempted to give a more informative measure of coverage for the different variables and have provided estimates based on what was submitted on the returns by active companies in each year.
   
5. When a company submits their CT600 return they may not complete all box items. Some may not apply to their circumstances, others omitted or set to zero. Layered upon this are a range of customer behaviours embodied in their return responses but not openly stated (for example, deliberate under or over declaration, or simple errors). The CSAT box item tables – that show CT600 return figures for any company reporting at least one box item at some point in the last 7 years – simplify this picture, showing reported values over zero and the remaining cells defaulted to zero. It could equally be set to blank, but either way, represents that no positive value has been reported. For this analysis, all zeros have been counted in the coverage statistics whereas null values have not.
   
6. Employee variables - this includes employee count, expenditure on wages, and average employee pay. These are sourced from Pay As You Earn (PAYE) Real Time Information (RTI) data and so a company would only report data here if they are PAYE registered. A company made up of just a number of directors, for example, would not report employee variables despite being an active company.

6.4 HMRC Research and Development Tax Credits

The R&D data are sourced from Corporation Tax returns for account periods ending in the specified financial years. Companies may have accounting periods beginning and ending at any point in time. For example, R&D expenditure pertaining to a company with an accounting period beginning in March 2017 and ending in April 2018 will be included in the 2018-2019 financial year, as will the expenditure for a company with an April 2018 to March 2019 accounting period.

Unlike the CSAT data, the R&D tax credit claims data is not annualised. This means that variation in account period lengths can result in multiple accounting periods, and therefore multiple tax credit claims in what appears to be the same financial year. However, companies may also make multiple claims in the same accounting period, e.g. if they are eligible for different R&D relief schemes.

There are 2 Corporation Tax Credit (CTC) schemes under which companies may claim tax credits subject to their eligibility: the Research and Development Expenditure Credit (RDEC) and Small or Medium-sized Enterprise (SME) schemes. Irrespective of the scheme, companies receive tax credits (reliefs) through 1 of 2 mechanisms – Corporation Tax liability deductions or directly payable credit. The tax credit (relief) amount is determined by the company’s qualifying expenditure.

The R&D variables used in the analysis were:

• Number of claimants. This is determined by the number of matches between the company lists and the R&D dataset. If a company has not been matched to the R&D dataset, it is deemed to be a non-claimant (either because it has not submitted a tax return, or because its tax return does not include an R&D relief claim).
  
• Qualifying expenditure. This is an estimate of the company’s investment in R&D activity for the R&D tax relief claim in question. Companies are not required to report this, and report only the amount of relief being claimed. The estimated qualifying expenditure is reverse engineered from the claim amount using the rules of the R&D tax relief scheme.
  
• Additional government support. The R&D tax relief schemes are deemed to be a form of additional government support. While HMRC has a time-series of the R&D data, data on other forms of additional government support is less readily available. DSIT supplemented HMRC’s R&D data with snapshots of data from the British Business Bank, Innovate UK and other sources. Since only snapshots of the latter sources were available, they were compared to a snapshot of the R&D data (the most recent year available).

6.5 Data timings

The analysis uses tax data from Corporation Tax (CT) and Pay As You Earn (PAYE). To correctly interpret the findings in this publication, it is important to understand fundamental structural differences between these 2 taxes.

Corporation Tax is paid by incorporated companies, who can determine their own accounting periods and do not have a set ‘tax year’. By contrast, PAYE is paid by employers, with a tax year from 6 April to 5 April the following year.

Thus, estimates based on PAYE data (such as employment and wages) are shown by ‘tax year’ while estimates based on CT data (such as turnover, profit, R&D and others) are shown on a ‘CT600 basis’, i.e. figures for a given financial year relate to company accounting periods ending in that financial year. This should be borne in mind when comparing estimates based on different tax data, as well as interpreting computed variables such as GVA, which rely on variables from both.

As the diagram below illustrates, for some companies (depending on their CT accounting period), the timing of the economic activity reported for CT and PAYE could differ by up to a year (with CT lagging behind PAYE) even though both periods would appear under a single financial year in the analysis in this report. This is standard practice for reporting CT statistics, which are not typically compared with PAYE statistics, which reflects the novel nature of this analysis.

Furthermore, the company lists which identify the 3 the tech sectors are snapshots, and variation in the sector population over time is not captured in the time series shown. The Eng Bio snapshot was finalised in October 2023 , the Quantum snapshot in April 2024 and the RAS snapshot in September 2024.

The diagram below illustrates all 3 points on data timing described above.

Figure 63: Timings for Corporation Tax (CT) and PAYE data availability and extraction, and DSIT company list snapshots

6.6 Sector identification

For the purposes of developing company lists using the Data City, each sector required a taxonomy to define the sectors. These are outlined below.

6.6.1 Eng Bio

Taxonomy

The goal for the DSIT team in summer 2023 when developing the Eng Bio RTIC was to build out an exhaustive taxonomy of application focused Eng Bio companies, but also those in the enabling Eng Bio supply chain.

The taxonomy shown below includes the definitions of each sub-sector, for which an individual company list has been developed to capture companies with a UK presence (except for the subsectors with *, where there were either insufficient numbers of companies or inherent limitations in the ability to create an accurate list due to the detail of website text.)

It should be noted that several EB companies are counted as operating in different subsectors (16% of firms are in more than one sub-sector), however when aggregating up to a sector or the overall RTIC companies in multiple sub-sectors are only counted once. Furthermore, the overall RTIC contains both parent companies and subsidiaries, unless it is flagged as “grouping claims at URL level” to remove this duplication.

Engineering Biology: Application, * = not included in companies list

• Chemicals and Materials Using engineering biology to produce chemicals and materials for purposes outside of healthcare and farming.

• High-value compounds
  Fermenting engineered organisms to create high value chemicals. This including cleaning products, energetic materials, functionalised materials, butanol, acetone, petrochemicals.

• Materials such as biobased/degradable products or sustainable fashion.
  Biobased & biodegradable material products, such as environmentally friendly and novel ways to produce dyes, silk, and packaging etc.

• Biosensing*
  Detecting molecules using living systems/organic technology inspired by living systems.

• DNA for data storage*
  This method of data storage is still in the experimental stages and is not yet practical for everyday use. However, it has the potential to provide a long-term, stable, and dense storage medium that could be particularly useful for archiving large amounts of data.

• Biofuels and hydrogen Engineering biology to produce fuels

• Fuels, such as Jet fuel / Biofuels / Green Hydrogen
  Engineering biology systems for producing chemical fuels.

• Environment & CO2 Capture Engineering biology for environmentally friendly purposes. Whether that is removing carbon from the atmosphere or industrial processes, or removing toxins from the environment.

• Bioremediation/Water treatment.
  Employing engineering biology to remove contaminant, pollutants and toxins from water and soil.

• Bio Energy and Carbon Capture.
  Engineering biology for creating electricity, and for converting carbon dioxide or other gaseous carbon compounds into products.

• Health and Life Sciences Engineering biology for the healthcare industry, human or veterinary.
  [This list was trained at this level, rather than at the sub-sector level due to too much overlap in website text]

• Antibiotics / small molecules / Physiological monitoring (traditional approaches made/disrupted in new way)*
  Using engineering biology to produce small molecules and methods for administering medicinal chemicals.

• Monoclonal Antibodies / antibodies / T-cells / mRNA / vaccines*
  Products for the immune system of organisms. Human or animal. Traditional vaccine techniques, mRNA vaccines, monoclonal antibodies, antibodies, T-cells, and anything that is given to provide immunity or improve the immune response to a disease and is not personalised to the individual.

• Targeted, personalised, and biotherapies utilising organic systems*
  Personal medicine, bespoke therapies targeted to the individual. Can include synthetic organs, and live biotherapeutic products. Phage therapies have been included within this group as despite not being necessarily personalised, they use incredibly specific targets and harness phages which are organic.

• Agriculture and food Engineering biology utilised to produce the things we eat

• Sustainable Agriculture / Eco-Friendly Agrochemicals.
  Engineering biology to create products for the agricultural sector. Examples include: precision bred crops, sustainable fertilisers, and alternative pest control methods.

• Foods (such as alternative meat/protein/fat).
  Using engineering biology to create foods–such as cultured meats, alternative proteins and fats–that offer an environmentally friendly alternative to traditional farming methods.

Engineering Biology: Supply Chain, * = not included in companies list

• Physical Assets Companies hardware-related products and services to engineering biology firms

• Small Scale Manufacturing
  All hardware needed for proof of concept, from pipettes, glassware, benchtop centrifuges, through to autoclaves and automated platforms such as liquid handling robots. This also includes companies that construct lab space

• Pilot / Mass Manufacturing
  The equipment and skills needed for running pilots and proof of scalability for engineering biology services and products. The infrastructure and the skills needed to construct and maintain the equipment required to produce engineering biology services and products at commercial scale (e.g. bioreactors >100 kL). This includes the companies that construct these factories and facilities.

• Biological materials and reagents Companies providing “biological materials and reagent”-related products and services to engineering biology firms.

• Biological materials and reagents
  Pre-processed intermediate commodities. This includes enzymes, chemicals, biological chassis, strains, and media supplements.

• Feedstocks*
  The largely unprocessed primary commodities and processed primary commodities for media. This includes biomass.

• Nucleotide synthesis and sequencing (also an application)
  The equipment and suppliers for DNA sequencing and synthesis, as well as of other nucleotides.

• Diagnostics The equipment for diagnostics including for quality assurance and control

• Diagnostics This equipment provides the evidence and credibility needed at every stage of an engineering biology companies existence. While larger companies will have much of this inhouse for security, smaller companies will rely on using diagnostics capabilities present in universities, research institutions, and CROs.Examples of diagnostic equipment and processes include, but are not limited to, NMR, HPLC, LCMS, X-ray crystallography, enzyme kinetics, assay development, optimisation, troubleshooting, analysis of materials, and quality control.

• Computational Companies providing computational-related services and products to engineering biology firms

• Supercomputing
  The hardware for processing, data storage. Such as GPUs, High performance computer clusters, and servers.

• AI, Bioinformatics and omics
  The software and data used for bioinformatics, omics, and any other program required for your work from simple scripts through to machine learning platforms

• Robotics
  The hardware for physical interaction with processes.

• Humans Companies providing human capital / contract services to engineering biology firms.

• Consultancies, Recruitment and Business Skills*
  Consulting and recruitment businesses specialise in providing expert advice and sourcing talent for companies in the engineering biology sector. They assess client needs, offer strategic solutions, and identify skilled professionals to fill specific roles. Leveraging industry knowledge and networks, these firms bridge the gap between companies and potential employees, enhancing organizational efficiency and growth. Their services cater to diverse sectors, adapting to unique market demands and trends.

• Contract Research Organisations* Companies which act similarly to consultants but have lab facilities and capabilities to take on extra work from other companies to perform specific research. Massively reducing costs compared to building facilities and retaining skills in house for other companies.

• Contract Manufacturing Organisations*
  Companies which act similarly to consultants but have lab facilities and capabilities to take on extra work from other companies to perform manufacturing commissions. Massively reducing costs compared to building facilities and retaining skills in house for other companies.

6.6.2 Quantum

The Quantum Technology RTIC is intended to cover all companies with a presence in the UK involved in quantum computing, quantum communication, quantum sensing and/or quantum materials. This RTIC has been developed through a collaboration between DSIT internal analysts and subject matter experts within Innovate UK Business Connect and the Government Office for Science (GOS) over the first quarter of 2024. Definitions were based on the GOS taxonomy and collaborative quality assurance processes performed with The Data City.

The Quantum Technologies landscape is diverse and involves many actors. Criteria for inclusion within this RTIC takes consideration of companies selling and/or developing products or services that exploit quantum phenomena and companies selling and/or developing products or services whose principal applications are within quantum technologies. As this is a rapidly changing area it should be noted that this RTIC does not assess the whole quantum ecosystem and is designed to minimise the risk of scope creep. Defining the sector in this way allows us to develop estimates of the economic contribution of the quantum sector with greater confidence.

To understand the taxonomy or the caveats associated with the RTIC in more detail, please get in touch with the DSIT Office for Quantum team or with The Data City. The current RTIC will have future updates, if and when there are new sources that are brought to the attention of the DSIT Office for Quantum team or, DSIT deem the list to need a substantive update due to time elapsed or if there are new subsectors to include.

Quantum Computing
Companies involved in hardware (e.g. qubit technologies, quantum processors) and/or software (e.g. quantum algorithms, programming languages, and simulation tools) to process information and/or perform computations based on the use of quantum bits (qubits) or principles of quantum mechanics to inform the design of quantum computers.

Quantum Communication
Companies involved in quantum-enabled technologies for secure communication, including the use quantum effects to transmit, encrypt, or decrypt classical information (e.g. Quantum Key Distribution), and/or the transmission of quantum information across networks (e.g. Quantum networks).

Quantum Sensing
Companies involved in technologies that use the principles of quantum mechanics to sense and/or image attributes (e.g. physical, chemical, biological) operating at the atomic level allowing for high sensitivity beyond classical devices.

Quantum Materials
Companies involved in materials whose properties are primarily defined and governed by quantum mechanical effects at macroscopic scales and with applicability to quantum technologies

6.6.3 Robotics and Autonomous Systems

Taxonomy
The team prioritised areas in which robotics and automation companies apply their technologies. This enabled the machine learning tool to recognise terminology relevant to application areas. Whilst several robotics companies operate in different sectors, the team acknowledge this overlap. Companies are only counted once in the overall RTIC, although subsidiaries of the same parent company may be included.
 
The taxonomy shown below includes definitions of each sub-sector, for which an individual company list has been developed. This taxonomy is not intended to be mutually exclusive but does aim to be collectively exhaustive.

Rationale for taxonomy
The taxonomy is based on a decomposition of operating domains where robotics and autonomous systems are currently applied. The RTIC tool uses language and word similarity on company websites to compare and find new companies to add to each list. This drives the second layer decomposition by singling out areas which are likely to use similar terminology to describe themselves. So, for example, consumer applications decompose into 5 market sectors each with distinct identity; a cleaning company is unlikely to use the same phraseology as a company supplying into the hospitality sector. 
In choosing the market sectors care was taken to avoid over dividing the categories where this resulted in lists that were too specific, resulting in only a handful of companies. On occasion this was unavoidable because the language used by companies was highly market specific, for example in rehabilitation or recycling. The decision to keep small lists was based on the difficulty encountered in accurately including those companies into more broadly defined lists.

In addition to these end user market sectors the RTIC lists were extended to include the robotics supply chain by creating lists for component suppliers, integrators and AI for robotics software suppliers. This was done because these companies typically operate cross-sector, or across a set of sectors, and usually include phraseology from all the sectors they operate in. This makes them difficult to distinguish from sector specific companies in the lists so separate lists for the UK based robotics and autonomous systems supply chain were created to improve monitoring of these companies. 

Over time, as the market for robotics and autonomous systems expands, it may become necessary to further decompose some of the high-level categories; for example, the Circular Economy category within manufacturing might need to split into domestic and industrial markets, or Construction split between domestic building and infrastructure building. This will need to be monitored on a regular basis. 

Robotics and Autonomous Systems

Mobility/Transport
Robotics used to convey goods or people using public spaces or over long distances within private spaces.

• Unmanned Aerial Systems / Drones
  Robots that fly, whether operating in general airspace or in enclosed spaces. Covers drone suppliers, suppliers of drone components and professional operators as well as those delivering services based on drone technology.
• Maritime
  Robots operating on or under the water, either at sea or within inland waterways. Covers robot suppliers, operators and those delivering professional services based on robots working on or under water.
• CAV / Self-driving vehicles
  Land based robots that carry people or goods either on the national road network, or on private road networks or off-road. Covers vehicle suppliers and related professional service providers.

Logistics / warehousing
Mobile robots operating within warehouses and in logistics operations, e.g. in factories, supermarkets, delivery centres etc. Covers robot suppliers, whole system suppliers and warehouse operators with robotics as a core capability.

Agriculture
Robots used in all aspects of agriculture, horticulture and farm animal management. Covers suppliers and specialist robot focused agri-tech installers and related robotics service providers.

• Milking / animal
  Robot systems used to manage and tend animal livestock covering all aspects of animal husbandry and the supply of robot specific services.
• Weeding, seeding and Harvesting
  Robot systems used to manage arable land and crops including pest control and fertilisation. Covers all aspects of the growing cycle and robot-based services relevant to crop management.
• Vertical Farming
  Robot systems used in all aspects of crop production in vertical farms. Covers equipment supply and robot related service provision.

Healthcare
Robotics used in all aspects of healthcare provision including all aspects of healthcare delivery within health facilities.

• Surgical
  Robots used in invasive procedures with varying degrees of autonomy or guided by a clinician. Covers surgical procedures and invasive sample taking.
• Pharma (lab robotics)
  Robots and automation used in pharmaceutical laboratories, in sample testing laboratories, pharma industry manufacturing and in pharmacy logistics within a clinical environment.
• Rehab
  Robots used to provide rehabilitation and patient assessment services. Includes exoskeletons and rehabilitation services for recovery from surgery and trauma
• Assistive
  Robots used to provide assistance to patients or assistance for clinical staff in interacting with patients. Covers both at home and clinical environments.
• Diagnostic
  Robots used in non-surgical procedures covering non-invasive sample taking, internal and external diagnostics and patient positioning for diagnostics.

Security
Robots used in any military-based use case across land, sea and air. Covers all aspects from logistics to operational use and includes military related service provision relevant to military robotics.

Infrastructure
Robots used in all aspects of the built infrastructure.

• Inspection and Maintenance
  Robots used in the inspection and maintenance of infrastructure in transport, energy and industry. Covers the inspection and maintenance of bridges, tunnels, chimneys, pipes, cables etc. and the provision of robotics related services. Includes facilities on land and at sea.
• Construction
  Robots used in the construction of infrastructure, on land, underground, on sea and undersea. Covers building and demolition, preparation of building materials, land clearance and preparation for construction, and construction of bridges, tunnels and other elements of transport, industry and energy infrastructure.

Manufacturing
Robotics used in the manufacture of goods including the manufacture of parts, sub assembles and full product assembly. Robotics used in the manufacturing of goods, including food, and the production of materials. Includes circular economy.

Packaging
Robots and automation used in packaging, particularly focusing on fast moving consumer goods.

Energy
Robotics used in the generation and supply of energy.

• Nuclear
  Robotics and automation used in the nuclear supply chain covering development, operation, maintenance, decommissioning and waste management activities. Covers robotics related service provision and relevant inspection and monitoring support services.
• Renewables
  Robotics used in the production of renewable energy either land or sea based. Including robotics related service provision and the development of renewable energy infrastructure.
• Oil Gas, Grid & Other
  Robotics and automation used in the production and distribution of oil and gas products including the services related to oil and gas infrastructure including inspection, drilling, extraction, processing and decommissioning.

Recycling
Robotics and automation used in waste management and recycling. Covering sorting machinery and robotic systems for the recycling industry.

AI software
Covers specialised AI service and module provision related to Robotics and Robotic operations.

Integrators
Covers the service providers around the integration of robotics into end user environments across all sectors.

RAS Component suppliers
Robotics component and sub-system suppliers, technology developers and robotics industry service suppliers including robotics manufacturing services.

Space
Robotics and automation used within the space sector either in orbit or in planetary exploration. Covers manufacture of space infrastructure and vehicles and the space-based servicing and delivery of services using robotic systems.

Consumer
Robotics used within a consumer setting and in public spaces.

• Mowing
  Robotics and automation used to provide domestic assistance in everyday tasks.

• Cleaning Robotics and automation used to provide domestic assistance in everyday tasks.

• Retail and Hospitality
  Robotics used in hospitality and retail, in shops, malls, retail parks, hotels, conference and exhibition venues. Covering cleaning, delivery, advertising, marketing, promotions, guest relations, hotel services, navigation services and stock monitoring and control.

• Sports and Entertainment
  Robotics used in sports and entertainment services. Covering sports and recreational activities and the use of robotics in entertainment venues.

• Educational
  Robotics used for educational purposes and in training.

6.7 Loss-making companies

HMRC data on companies in EngBio and Quantum demonstrate the negative returns of some SME companies operating in these sectors.

These losses for SMEs exist because the method used to produce these sector lists captures companies focused on these tech areas that are early in their development stage, often pre-revenue and pre-profit. This negative return is common for younger, smaller firms, but will be more pronounced in this dataset compared to commercially available data sources because of the higher variable coverage for turnover/PGTP, there are also more negative values of the PGTP variable (where in commercially available datasets these negative values are less likely to be reported).

Any interpretation for what these negative profits mean for the sector should consider 2 points:

1. Large firms are driving positive returns within these sectors.
   
2. Even if the large company profits outweigh the SME losses in aggregate, it is hard to infer whether the overall sector is profitable or has positive GVA.

This challenge in taking the overall sectoral view stems from an apportionment problem: larger companies tend to be more diversified, and only a portion of their activities falls within the identified subsector. Because there is no systematic way to split out and attribute these activities, we cannot robustly determine the sector’s overall position.

To explain this issue visually, consider the below typical “J-Curve” trajectory for an emerging technology firm, where there is a pre-revenue period during commercialisation followed by an eventual profit. We know that the number of firms in category A are > B when we look at the HMRC SME companies aggregate PGTP figure. However, this does not tell us whether A + B + C are > 0 (or not), because of the above mentioned apportionment issue (which is more acute in area C than elsewhere).

Figure 64: J Curve for emerging sectors

1. https://thedatacity.com/rtics/ ↩

2. beauhurst.com ↩