Evaluation of Tax-advantaged Venture Capital Schemes (2018) – executive summary

1. Glossary 

Term	Meaning
Difference-in-differences (DiD)	A statistical technique that calculates the effect of a treatment (ie an explanatory variable or an independent variable) on an outcome (ie a response variable or dependent variable). It compares the average change over time in the outcome variable for the treatment group, with the average change over time for a control group.
Knowledge intensive companies (KICs)	Knowledge intensive companies are those carrying out research and development or innovation activities, and meet certain conditions.
Ordinary shares	Ordinary shares entitle the shareholder to one vote per share, to equal participation in dividends and, if the investee company is liquidated, to a share of the proceeds of the company’s assets after all the debts have been paid.
Small and medium-sized enterprises (SME)	Businesses whose staff, turnover and/or assets numbers fall below defined limits. The EU defines them as businesses that employ fewer than 250 persons and with an annual turnover not exceeding 50 million euro, and/or an annual balance sheet total not exceeding 43 million euro. In this research, SMEs are defined as businesses with fewer than 250 employees.In the analysis, SMEs are divided into the following groups: ‘micro’ (0-9 employees), ‘small’ (10-49 employees) and ‘medium’ (50-249 employees).

2. Executive summary 

NOTE FOR THE READER: The following is an executive summary report of an evaluation of the EIS and VCT scheme which was completed in 2018. A more recent evaluation of these schemes has been completed and aimed to address some of the National Audit Offices criticisms of this evaluation. The more recent evaluation has been published alongside this report on GOV.UK. 

2.1 Introduction 

The Enterprise Investment Scheme (EIS) and Venture Capital Trust (VCT) scheme are policy interventions that aim to support the growth of newer small and medium-sized enterprises (SMEs). SMEs, new start-ups and companies focused on innovation often struggle to access sufficient finance. The schemes, sometimes referred to as “venture capital or VC schemes”, provide tax incentives to private individuals to encourage them to invest in companies and social enterprises that are not listed on any  recognised stock exchange. They can invest directly in companies that qualify for the EIS, or in financial entities that pool investments across companies that qualify for the VCT scheme. 

In 2016 to 2017, VCTs issued shares to the value of £570 million and HMRC estimates suggest a cost of £251 million in foregone tax. For the £1.8 billion of EIS Investments in 2016 to 2017, the costs were estimated at £648 million. 

The European Commission (EC) approved the EIS and VCT schemes as state aid in 2009. As part of a 2015 amendment, which approved the schemes to 2025, the EC noted the UK’s commitment to providing an independent evaluation, as required under the Risk Finance Guidelines (RFGs). This analysis was commissioned by HMRC for that purpose. This independent evaluation of EIS and VCT considered the following questions, with respect to the schemes: 

1.  do they improve access to risk finance for SMEs 

2.  do they contribute to the development of a more efficient market for risk finance 

3.  do they incentivise new and additional investment by private investors 

4.  do they ensure other sources of private investment are not crowded out 

5.  does this State aid replace the normal business risk 

6.  to what extent, if any, does this State aid distort competition 

7.  do they have any indirect impacts, and if so what are their effects 

This report addresses the effect of investments made under the VC schemes in the period from 2009 to 2010 to 2014 to 2015, and reflects the rules applied by HMRC in this period. It is important to note that after the period of study the rules on acceptable investments changed significantly, most notably through the exclusion of VCTs from being allowed to invest in management buyouts since November 2015, and the introduction from 2018 of a new ‘risk to capital’ condition. These changes can be seen as a response to some of the behaviours uncovered in this study that were apparent between 2009 to 2010 and 2014 to 2015; meaning that some of the problems identified in this report, may already have been addressed. 

2.2 Market Failure 

Supply side market failure occurs when investors decide not to pursue a proposed investment, based on a lack of information on the company, rather than the merit of the idea itself. For instance, a majority of companies reported that an investment would not have happened without EIS or VCT finance (40% definitely not and 28% probably not), implying that the schemes supported valid investment opportunities that would not have been taken up without them. 

Demand side market failure occurs when a company lacks the expertise to present a compelling investment proposition. EIS investors’ main investment was most likely to be in small, start-up companies, where we could expect both experience of the management team (demand side) and the problems of information asymmetry (supply side) to be most pronounced. Survey findings support this. For instance: 

• around half of EIS investors (54%) said that their main investment was in a company that was less than 2 years old 

• 31% invested in ‘early stage companies’ (established for between 2 to 5 years) 

• 6% invested in established companies (established for five years or more) 

• 40% invested in a company with fewer than 10 employees 

Unregulated free markets will likely under-produce innovation because the positive externalities or ‘spillovers’ associated with these activities (for instance, new technologies that benefit all society); are not valued by individual investors and companies in their decisions. The main reasons for seeking external finance, and new products and services that companies had introduced, suggested that EIS and VCT were supporting activities that had the potential for positive externalities and/or were associated with high levels of risk. For instance: 

• slightly less than half of companies (48%) reported new technology because of the finance 

• slightly less than two-thirds (64%) reported new products or services because of the finance 

• 29% of investee companies were ‘knowledge intensive’ companies (KICs) under the EU definition 

There is debate over the exact nature and extent of market failures in SME finance, and a wider debate on the role of smaller companies in ensuring effective working of competitive markets. But there is broad agreement that market failures are most pronounced for: 

• companies at an early stage of development (especially new start-ups) 

• smaller businesses within the broader ‘SME’ grouping 

• innovative companies that are often associated with higher levels of risk and/or intangible assets 

This study gathered evidence to assess the extent to which the EIS and VCT schemes were seen to support these types of companies. 

2.3 Research methodology 

This research involved an econometric evaluation of the EIS and VCT schemes to capture causal impacts. Descriptive analysis informed discussion of market failure and the extent to which the schemes could produce distortionary impacts on markets. Surveys of investors and investee companies captured a range of additional measures to complement analysis. There were 3 main components: 

• telephone surveys of investee companies and investors in EIS and VCT schemes 

• interviews took up to 20 minutes and were conducted in spring 2018 (the ‘treatment’ groups) 

• 437 investee companies (42 VCT-only, 379 EIS-only and 16 that received both) and 805 investors (332 VCT-only, 300 EIS-only and 173 that invested in both) were interviewed 

• concurrent telephone surveys with matched comparison groups of companies and investors (the ‘comparison’ or ‘control’ groups) 

• 400 companies and 401 investors made up the respective comparison groups 

• analysis of administrative data and broader descriptive statistics from desk research 

• in addition to difference-in-differences analysis (DiD), a panel data econometric approach evaluated the effectiveness of tax-advantaged investments, providing new insights into scheme impacts 

The survey data provided a rich set of indicators on which to compare EIS and VCT participants with respective comparison groups. The administrative data analysis captured impacts by comparing a larger number of representative EIS and VCT investees. Analyses of survey data provided a deeper understanding of issues, through for instance analysis of attitudinal indicators. Analyses of administrative data used econometric techniques more advanced than DiD, such as dynamic panel estimation. This dual data collection and analytic approach ensured that the various econometric analyses complemented each other, filling in evidence gaps from previous investigations. 

2.4 Conclusions  

The survey of investee companies confirmed that many companies supported by the schemes between 2009 to 2010 and 2014 to 2015, were those most likely to experience market failures. However, the results suggested that the VCT scheme focused less on companies where market failures predominate, compared with EIS. For instance, 84% of surveyed EIS companies were ‘completely new independent start-ups’ compared with 52% for VCT, and whilst 10% of VCT companies were micro enterprises, for EIS the figure was 44%. 

The survey provided evidence that the April 2011 changes to EIS significantly improved access to risk finance for SMEs. EIS had a positive and significant impact on a variety of company outcomes. This gave some reassurance that any significant impact of EIS, in terms of additional finance secured by surveyed companies, was being used to support activities that were the target of policy. 

The econometric evidence also suggested that the EIS scheme incentivised new and additional investment by private investors. Impacts identified at the extensive margin would be those that impact the number of investors engaged in the scheme. The intensive margin in this context relates to variation in the amount invested by a given set of investors. The analysis of investee company data captured impacts at both the intensive and extensive margin. 

The findings for VCT investors and investee companies were less positive than those for EIS. In many cases we were unable to find definitive evidence on VCT due to data limitations that put constraints on the analyses. However, in one area where robust analysis was possible, the findings from administrative data suggested that VCT had a significant negative impact on levels of finance flowing to companies. Companies that were the target of VCT, experienced significantly lower assets in the year after investment, when compared with an appropriate counterfactual. As a result, the study found evidence that VCT did not improve access to risk finance for SMEs and there was no evidence that VCT incentivised new and additional investment by private investors. 

As suggested in a variety of studies (for instance, Colahan et. al., 2016), both investor and investee survey respondents were less likely to include those using EIS and VCT schemes for tax mitigation purposes. Despite this, it is reassuring that EIS survey data included many companies that experienced market failure; as the study can conclude that when the scheme was correctly targeted, EIS significantly improved levels of finance, and this translated into significantly improved outcomes. By supporting companies that suffer from a variety of market failures, a significant proportion of EIS funding has contributed to the development of a more efficient market for risk finance. 

The magnitude of the positive impact on efficiency in the market for risk finance was determined by the proportion of ‘appropriate’ companies supported by the scheme in the wider EIS population. Due to data limitations (especially relating to industry sector) it was not possible to estimate this proportion. However, the findings from administrative data analysis, coupled with an understanding of the ways that EIS and VCT were used for tax mitigation during this period, provided insight. During the period of evaluation, ‘technology’ (incl. information and communication), ‘energy’ and ‘environment’ were the top 3 sectors receiving EIS and VCT investment. 

In the case of EIS, the study considered the example of solar farms, set up with investment support for the purchase of land and solar panels. Between 2009 to 2010 and 2014 to 2015 these companies were operated (at low risk) to provide an enhanced return because of the associated tax breaks. From the econometric analysis it is not possible to identify whether the extent of this activity had a material impact on overall sector-wide outcomes. Assets would be expected to rise when companies suffering from market failure were supported, and when EIS was being used solely for tax mitigation. The findings from econometric analysis of administrative data showed that, when considering the entire population of EIS companies between 2009 to 2010 and 2014 to 2015, there was a significant positive impact on levels of funding secured (as captured by the log of total company assets), when compared with the counterfactual. 

In the case of VCT, the study considered the example of management buyouts that do not represent an investment in ventures that can be considered risky, but allow investors to secure a return with generous tax breaks. Under this scenario (in contrast to EIS), it was not necessarily the case that higher levels of total company assets would be observed. Moreover, the panel data analysis identified a negative and significant impact of VCT on (log) total company assets. One explanation for this could be that the scheme was being used for tax mitigation purposes by some VCTs. 

If a significant proportion of VCT funding had flowed to companies that experienced market failure, this would be expected to dilute any findings arising from manipulation of the scheme, when analysing the whole population. The fact that this was not the case, suggested that between 2009 to 2010 and 2014 to 2015 the VCT scheme was not contributing to the development of a more efficient market for risk finance. Whilst some VCT investments contributed to the development of a more efficient market for risk finance, a significant proportion exerted a negative impact. 

It is likely that both EIS and VCT crowded out other sources of private investment between 2009 to 2010 and 2014 to 2015, but we could not gauge the magnitude of this. Some EIS and VCT funds were able to offer tax-advantaged returns, for investments that were not particularly high risk and this will have provided an advantage over competitors. These indirect impacts were likely to have affected product markets and possibly trade, where the investment in low risk SMEs gave recipient companies an advantage over their competitors. As some projects (whether EIS-supported solar farms or VCT-supported films) would have been supported anyway, then the provision of finance via EIS or VCT would likely have been on more favourable terms (as the investor would require less of a return from the investment, other things remaining equal), compared with the counterfactual, and this would provide an unfair competitive advantage. This is expected to have distorted competition in product markets (as well as in the market for finance), but the extent of such distortion was hard to estimate. It was likely to have been significant in the areas of ‘energy’ and ‘environment’ between 2009 to 2010 and 2014 to 2015. 

Overall, the findings from this research suggested that even during a period when there were concerns over the appropriate targeting of EIS and VCT, EIS was supporting companies subject to the relevant market failures and the scheme had a significant positive impact on the situation of these companies. Since then, it is expected that changes to the schemes will have better ensured that these impacts predominate, whilst distortionary impacts arising from tax mitigation motives, have reduced. Lessons have been learnt over this period, regarding the targeting of EIS and VCT, but it will be important to ensure that any future review of the schemes is robust. 

2.5 Data quality 

This study was modelled using pre-2015 data. Significant changes were made to the schemes in 2015 and 2018, and so future reviews will give a better sense of the benefits of the scheme, once the changes are embedded. The evaluation would be strengthened through richer data, for example, the collection of employee numbers. However, HMRC‘s ability to collect data beyond what is required for tax administration is limited. It is difficult to recommend a date for the forthcoming review, as this is dependent on data quality and the lag between company reporting and data collection. However, it would not seem likely that such a review could uncover useful additional insights before 2022.