Annex B. Methods and findings of the 50 individual evaluations

This Annex documents the evaluation studies included in the report. They are listed by the dominant policy focus of the programme evaluated, as per our 8 policy themes. The evaluations fall under the areas of Finance; Business Advice, Coaching, Mentoring and Counselling; Internationalisation; Innovation; Enterprise Skills and Culture; Inclusive Entrepreneurship; Regional and Local Focus; and Support in Areas of Disadvantage. There are no examples of Cluster evaluations.

Finance

Table B.1. The impact of government financial assistance on the performance and financing of Australian SMEs
TABLE REFERENCE	B1
PROGRAMME NAME	Direct financial assistance from the Australian government, including grants, subsidies and rebates.
DATES	Years when the programme was operating: 2005-2010 Evaluation period: 2005-2010 Year of the report: 2017 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 1 The programmes of direct financial assistance (subsidies, grants and rebates) aimed to improve access to finance for SMEs, to mitigate the effects of the financial crisis and to enhance the competitiveness of the supported firms.
	No	Improved access to finance, increased competitiveness
EVALUATION THEME	Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	SMEs in Australia
SOURCE OF EVIDENCE	Academic article: (Xiang and Worthington, 2017[2]). The impact of government financial assistance on the performance and financing of Australian SMEs. Accounting Research Journal, 30(4), 447-464. Available at:https://doi.org/10.1108/ARJ-04-2014-0034
COUNTRY	Australia
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Sales, profit, probability of obtaining other funding
NON-SURVIVORS INCLUDED?	No
DATA SOURCES	Survey data and accounting data: Data from the Business Longitudinal Database compiled by the Australian Bureau of Statistics combined with official accounting data. 508 supported firms (18.6% of surveyed firms) and non-supported firms from the survey.
STEP LEVEL	5
METHODS	Panel data approach Random effects regressions The authors estimate firm-level effects in t+1
EVALUATION QUALITY SCORE	2
RELIABILITY COMMENTS	Analysis tackles the issue of industry bias and area bias only partially. Although the authors use pre-intervention financial data and firm-level characteristics, the two-step estimation is not used as a methodological approach. Moreover, different kinds of interventions are combined in pooled results and we do not know any details about the outcomes of the individual programmes.
KEY FINDINGS	The authors find positive effects of the governmental support on sales, profit and on obtaining other funding.
PROGRAMME EXPENDITURE	N/A
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	N/A

Table B.2. The effectiveness of investment subsidies: evidence from a regression discontinuity design
TABLE REFERENCE	B2
PROGRAMME NAME	Flemish government Entrepreneurship Agency programme
DATES	Years when the programme was operating: 2004-2009 Evaluation period: 2001-2012 Year of the report: 2016 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 1 The programme aimed to stimulate economic growth through investment subsidies allocated to firms.
	Yes	Increased investments, increased competitiveness
EVALUATION THEME	Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	SMEs in Flanders region
SOURCE OF EVIDENCE	Academic article: (Decramer and Vanormelingen, 2016[3]). The effectiveness of investment subsidies: evidence from a regression discontinuity design. Small Business Economics, 47(4), 1007-1032. Available at: https://doi.org/10.1007/s11187-016-9749-2
COUNTRY	Belgium
REGIONAL/LOCAL	One region study
PERFORMANCE METRICS	Employment, fixed assets, sales, value-added, labour productivity and total factor productivity (TFP)
NON-SURVIVORS INCLUDED?	No.
DATA SOURCES	Commercial data and administrative data: Firm-level data from Bel-First database (Bel-First) in combination with data from Flemish government´s Entrepreneurship Agency 932 firms supported during 2004-2009 in Flanders region (2,966 supported in total), 4,463 non-supported firms from Flanders region (rejected applicants)
STEP LEVEL	6
METHODS	Panel data approach Regression discontinuity design The authors estimate firm-level effects from t+1 to t+3
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of area and selection bias. However, the number of SMEs included in the analysis is much lower compared to the population of supported enterprises.
KEY FINDINGS	The authors find positive effects of subsidies on fixed assets, employment, sales, value-added, labour productivity and TFP growth for very small firms, and they do not find any effects for larger firms.
PROGRAMME EXPENDITURE	250 million Euro allocated through the Entrepreneurship Agency (Agentschap Ondernemen) programme during 2004-2008 in Flanders region, Belgium. The authors estimate that the cost of one job created through subsidy was high (500 ths. EUR).
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The authors summarised the results in a policy paper published on the website of their research unit and disseminated them in some of the national newspapers. However, they have not presented the results to policymakers.

Table B.3. The economic impact of the Canada Small Business Financing Program
TABLE REFERENCE	B3
PROGRAMME NAME	Canada Small Business Financing Program (CSBFP)
DATES	Years when the programme was operating: 2004 Evaluation period: 2002-2006 Year of the report: 2012 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to facilitate access to finance for SMEs through the allocation of credit guarantees in cooperation with commercial banks in response to distortions on the financial markets.
	Yes	Improved access to finance, increased competitiveness
EVALUATION THEME	Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	New and established SMEs in Canada
SOURCE OF EVIDENCE	Academic article: (Chandler, 2012[4]). The economic impact of the Canada small business financing program. Small Business Economics, 39(1), 253-264. Available at: https://doi.org/10.1007/s11187-010-9302-7
COUNTRY	Canada
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Employment, revenues, profit and wages
NON-SURVIVORS INCLUDED?	No
DATA SOURCES	Survey data and administrative data: Data from the Survey on Financing of Small and Medium Entreprises (SFSME) combined with the official accounting data. Four groups of firms: 199 firms supported by credit guarantees in 2004 (11 000 loans guaranteed in 2004 in total), 121 rejected applicants, 621 approved applicants but non-supported firms, and 2 105 non-supported firms (randomly selected SMEs from SFSME).
STEP LEVEL	5
METHODS	Panel data approach OLS regressions in combination with a difference-in-differences (DiD) approach The authors estimate firm-level effects in t+2
EVALUATION QUALITY SCORE	3
RELIABILITY COMMENTS	Analysis tackles the issue of industry bias and area bias only partially. Although the authors use pre-intervention financial data and firm-level characteristics, the two-step estimation is not used as a methodological approach. Nevertheless, an asset of the study is combination of different kinds of control groups.
KEY FINDINGS	The authors find positive effects of the programme on salary, employment and revenues, but no significant effects on profit.
PROGRAMME EXPENDITURE	The total amount of public resources allocated through both programmes was approximately 30 mil. USD. in 2004.
MACRO IMPACT	The authors estimate that the approximately 11 000 loans guaranteed in 2004 have created 0.63 jobs each, meaning that the CSBFP created approximately 5 000 jobs. The authors calculate that each job created costs of about 6 000 USD.
POLICY IMPACT OF THE EVALUATION	The author did not present results to the policymakers.

Table B.4. Assessing the microeconomic effects of public subsidies on the performance of firms in the Czech food processing industry: A counterfactual impact evaluation
TABLE REFERENCE	B4
PROGRAMME NAME	Subsidies allocated within the Czech Operational Programme for Enterprises and Innovation (OPEI)
DATES	Years when the programme was operating: 2007-2013 Evaluation period: 2005-2015 Year of the report: 2019 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 1 The programme aimed to improve the competitiveness of the Czech firms through the allocation of investment subsidies.
	Yes	Increased competitiveness, increased investments
EVALUATION THEME	Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	SMEs in the Czech Republic, the evaluation focused on firms in the Food industry
SOURCE OF EVIDENCE	Academic article: (Dvouletý and Blažková, 2019[5]). Assessing the microeconomic effects of public subsidies on the performance of firms in the Czech food processing industry: A counterfactual impact evaluation. Agribusiness, 35(3), 394-422. Available at: https://doi.org/10.1002/agr.21582
COUNTRY	Czech Republic
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Price-cost margin, return on assets (ROA), assets turnover, value-added per labour costs, long-run risk, tangible fixed assets, labour costs, sales
NON-SURVIVORS INCLUDED?	No.
DATA SOURCES	Administrative and commercial data: Firm-level data from databases Magnus Web and SSV, in combination with data from the Czech Ministry of Industry and Trade. 143 firms supported by subsidies (203 firms were supported in the Food industry in total) and 604 non-supported firms from the Czech Food industry (non-applicants).
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) in combination with a difference-in-differences (DiD) approach The authors estimate firm-level effects in t+2
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of area and industry bias.
KEY FINDINGS	The authors find positive effects on price-cost margin, value-added per labour cost, growth of sales and growth of tangible assets.
PROGRAMME EXPENDITURE	The total amount of public resources allocated through subsidies to the firms in the Czech food industry was 86.4 mil. EUR.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The results were presented at the Czech Ministry of Industry and Trade for the relevant stakeholders and they promised to consider and incorporate recommendations on better data collection and better organisation of the programme, stressing the growth potential in the evaluation of project proposals.

Table B.5. Do firms supported by credit guarantee schemes report better financial results 2 years after the end of intervention?
TABLE REFERENCE	B5
PROGRAMME NAME	START and ZÁRUKA programmes
DATES	Years when the programme was operating: 2017-2013 Evaluation period: 2005-2015 Year of the report: 2019 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 Programme aimed to allocate financial capital to new and established SMEs in the Czech Republic.
	Yes	Increase in employment, higher competitiveness
EVALUATION THEME	Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	SMEs in the Czech Republic
SOURCE OF EVIDENCE	Academic article: (Dvouletý, Čadil and Mirošník, 2019[6]). Do Firms Supported by Credit Guarantee Schemes Report Better Financial Results 2 Years After the End of Intervention?. The BE Journal of Economic Analysis & Policy, 19(1), 2018005. Available at: https://doi.org/10.1515/bejeap-2018-0057
COUNTRY	Czech Republic
REGIONAL/LOCAL	One country study, focus on new and established SMEs
PERFORMANCE METRICS	Total assets, tangible fixed assets, personnel costs, sales, price-cost-margin (PCM), return on assets (ROA)
NON-SURVIVORS INCLUDED?	Yes, 17% of supported firms went out of business.
DATA SOURCES	Administrative and commercial data: Firm-level data from databases Magnus Web and SSV, in combination with data from the Czech Ministry of Industry and Trade. 530 firms supported by credit guarantees (85 firms were supported within Start and 2,011 firms within Záruka scheme in total) and 4,945 non-supported firms (selected randomly from business register)
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) in combination with a difference-in-differences (DiD) approach The authors estimate firm-level effects in t+2
EVALUATION QUALITY SCORE	5
RELIABILITY COMMENTS	Analysis tackles the issue of industry bias and area bias. The authors analyse altogether schemes Start and Záruka. Separate results across schemes are not reported.
KEY FINDINGS	The authors find only a positive change in tangible fixed assets for the programme participants and the effects for the remaining variables were not found to be statistically significant.
PROGRAMME EXPENDITURE	The total amount of public resources allocated through both programmes was 164 mil. EUR.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The results were presented at the Czech Ministry of Industry and Trade for the relevant stakeholders and they promised to consider and incorporate recommendations on better data collection and better organisation of the programme, stressing the growth potential in the evaluation of project proposals.

Table B.6. The role of financial support in SME and economic development in Estonia
TABLE REFERENCE	B6
PROGRAMME NAME	Various grants managed by the Enterprise Estonia (EAS) government agency, i.e. start-up and development grants, research and development (R&D) grant, development of knowledge and skills grants, technology investment grants, export grants
DATES	Years when the programme was operating: 2004-2009 Evaluation period: 2004-2010 Year of the report: 2013 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 1 The programme aimed to promote regional economic development in Estonia through the allocation of public subsidies to firms.
	Yes	Increase in number of start-ups, increase in innovation activities, increase in competitiveness, higher export, higher competitiveness
EVALUATION THEME	Finance; Innovation; Internationalisation
INTERVENTION TYPE	Hard
TARGET GROUPS	SMEs in Estonia
SOURCE OF EVIDENCE	Academic article: (Hartšenko and Sauga, 2013[7]). The role of financial support in SME and economic development in Estonia. Business & Economic Horizons, 9(2), 10-22. Available at: https://doi.org/10.15208/beh.2013.6
COUNTRY	Estonia
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Sales, labour productivity
NON-SURVIVORS INCLUDED?	No
DATA SOURCES	Administrative data: Data provided by Enterprise Estonia and Estonian Commercial Register. 508 firms supported during 2004-2009 by various grants (100% of population), 3,921 non-supported Estonian firms (non-applicants, 10% of all firms within sectors randomly selected)
STEP LEVEL	5
METHODS	Panel data approach Fixed and random effects regressions in combination with a difference-in-differences (DiD) approach The authors estimate firm-level effects from t+1
EVALUATION QUALITY SCORE	3
RELIABILITY COMMENTS	The analysis does not tackle the issue of area and industry bias. Control variables for firm-level observable characteristics are missing. The authors analyse altogether different kinds of grant schemes and programmes. Separate results across programmes are not reported.
KEY FINDINGS	The authors find positive effects on sales and labour productivity.
PROGRAMME EXPENDITURE	13.87 mil. Euro allocated through Enterprise Estonia (EAS) government Agency during 2004-2009 in Estonia.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	N/A

Table B.7. Impact evaluation of EU subsidies for economic development on the Hungarian SME sector
TABLE REFERENCE	B7
PROGRAMME NAME		Economic Development Operational Programme and Regional Development Operational Programmes from the EU Structural Funds and the Cohesion Fund
DATES		Years when the programme was operating: 2007-2013 Evaluation period: 2003-2015 Year of the report: 2017 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available		Yes	Objective specification score: 2 The programmes aimed to mitigate regional disparities through investment subsidies allocated to firms. Eligible activities included employment enhancement; support of research, development and innovation; environmental investments; development of production plants, technology and capacity; development of tourism; development of corporate information and communication technology; corporate consultancy; and they were supported through the direct subsidies and financial instruments.
		Yes	Increase in employment, higher competitiveness, higher innovation activity
EVALUATION THEME		Finance; Innovation
INTERVENTION TYPE		Hard
TARGET GROUPS		SMEs in Hungary
SOURCE OF EVIDENCE		Working paper: (Banai et al., 2017[8]). Impact evaluation of EU subsidies for economic development on the Hungarian SME sector. MNB Working Papers 8 (No. 2017/8). Available at: https://www.econstor.eu/bitstream/10419/189891/1/mnb-wp-2017-8.pdf.
COUNTRY		Hungary
REGIONAL/LOCAL		One country study
PERFORMANCE METRICS		Employment, value-added, sales, profit, tangible assets, labour productivity
NON-SURVIVORS INCLUDED?		No.
DATA SOURCES		Administrative data: Firm-level data were obtained from the National Tax, and Customs Administration (NTCA) and programme data come from the Unified Monitoring Information System (EMIR). Additional firm-level data were obtained from the Hungarian Central Statistical Office´s Business Register. 9 636 firms supported by subsidies (19 866 firms supported in total) and 2 587 firms supported by financial instruments (13 538 firms supported in total), 192 570 non-supported firms (non-applicants).
STEP LEVEL		6
METHODS		Panel data approach Propensity score matching (PSM) in combination with a difference-in-differences (DiD) approach, fixed-effects (FE) regressions The authors estimate firm-level effects from t+1 to t+4
EVALUATION QUALITY SCORE		4
RELIABILITY COMMENTS		Analysis tackles the issue of area and industry bias. A large sample and a complex evaluation study.
KEY FINDINGS		The authors find positive effects on employment, value-added, sales, profit, tangible assets, but insignificant effects on labour productivity. The separate results across firm-size (micro, small and medium) are provided as well as results across the programmes. The authors do not find differences between the outcomes of subsidies and financial instruments.
PROGRAMME EXPENDITURE		11 067 billion Hungarian Forints (HUF) was allocated through the programmes during years 2007-2013 (1$ = 221 HUF in November 2013).
MACRO IMPACT		N/A
POLICY IMPACT OF THE EVALUATION		The results were presented to the governmental agencies responsible for the allocation of EU subsidies (Prime Minister’s Office and the Ministry for National Economy) in 2017, and the authors were then asked for a consultation concerning the programming period 2021-2027.

Table B.8. Public credit guarantee schemes and SMEs’ profitability: Evidence from Italy
TABLE REFERENCE	B8
PROGRAMME NAME	Credit Guarantee Scheme Fondo Centrale di Garanzia (Central Guarantee Fund)
DATES	Years when the programme was operating: 2000- (nowadays) Evaluation period: 2005-2011 Year of the report: 2019 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to facilitate access to finance through the allocation of the credit guarantees in cooperation with commercial banks in response to distortions on the financial markets.
	Yes	Improved access to finance, increased competitiveness
EVALUATION THEME	Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	SMEs in Italy, focus on new and established SMEs
SOURCE OF EVIDENCE	Academic article: (Caselli et al., 2019[9]). Public Credit Guarantee Schemes and SMEs’ Profitability: Evidence from Italy. Journal of Small Business Management (forthcoming). Available at: https://doi.org/10.1111/jsbm.12509.
COUNTRY	Italy
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Return on investment (ROI)
NON-SURVIVORS INCLUDED?	No.
DATA SOURCES	Administrative and commercial data: Firm-level data from AIDA Bureau van Dijk, in combination with data from the Central Guarantee Fund. 15 562 firms supported by credit guarantees (55% of all supported firms) and 23 000 non-supported firms (collected from commercial database, i. e. non-applicants).
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) in combination with a difference-in-differences (DiD) approach The authors estimate firm-level effects from t+1 to t+2
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of industry bias and area bias. The authors provide estimates across firm size and industry.
KEY FINDINGS	The authors find positive outcomes of the programme on ROI, however only for micro and small firms. They further report negative impact in the case of medium-sized firms.
PROGRAMME EXPENDITURE	The total amount of public resources allocated to the scheme was 2 bil. EUR during 2008-2012.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The authors have presented results to the representatives of the Central Guarantee Fund and the Italian national promotion bank. However, the authors are not familiar about the specific changes, based on this evaluation report.

Table B.9. Are lending relationships beneficial or harmful for public credit guarantees? Evidence from Japan's Emergency Credit Guarantee Programme
TABLE REFERENCE	B9
PROGRAMME NAME	Japan’s Emergency Credit Guarantee (ECG) Programme
DATES	Years when the programme was operating: 2008-2011 Evaluation period: 2008-2009 Year of the report: 2013 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to mitigate the negative effects of the financial crisis and to improve access to finance for SMEs.
	Yes	Improved access to finance, higher competitiveness, mitigation of negative effects of the financial crisis
EVALUATION THEME	Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	SMEs in Japan
SOURCE OF EVIDENCE	Academic article: (Ono, Uesugi and Yasuda, 2013[10]). Are lending relationships beneficial or harmful for public credit guarantees? Evidence from Japan's Emergency Credit Guarantee Program. Journal of Financial stability, 9(2), 151-167. Available at: https://doi.org/10.1016/j.jfs.2013.01.005
COUNTRY	Japan
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Employment, loans obtained from a bank, interest payments, cash ratio, credit score, tangible fixed assets, sales, return on assets (ROA)
NON-SURVIVORS INCLUDED?	No.
DATA SOURCES	Administrative and survey data: Data obtained via a RIETI survey that was conducted by Research Institute of Economy, Trade and Industry, a research institution affiliated with the Ministry of Economy, Trade, commercial firm-level data from Nikkei Financial QUEST, data from Financial Services Agency, and other online sources. 365 firms supported by credit guarantees in 2008 and 2 134 non-supported firms (non-applicants, selected randomly from Tokyo Shoko Research database).
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) in combination with a difference-in-differences (DiD) approach The authors estimate firm-level effects t+1
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of industry bias and area bias. The authors provide interesting insights into firm-bank relationships by incorporating many bank-related variables.
KEY FINDINGS	The authors find that the programme significantly improved credit availability for supported firms. However, the authors could not find positive effects on profitability, investment and employment. On the contrary , they found a negative effect of the scheme on credit score of the supported firms.
PROGRAMME EXPENDITURE	The total amount of public resources allocated through the programme was 27.1 trillion yen (approximately 300 billion U.S. dollars).
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The authors presented the paper to the government representatives (SME Agency of the Ministry of Economy, Trade and Industry) and the central bank (Bank of Japan) officials on several occasions. Officials at the SME Agency collected academic research results on the effectiveness of the credit guarantee scheme and used them for drafting a policy package to reform Japan's credit guarantee programme. The reform started in 2018. The paper was also presented at the International Monetary Fund (IMF).

Table B.10. Evaluation of credit guarantee policy using propensity score matching
TABLE REFERENCE	B10
PROGRAMME NAME	Credit Guarantee Schemes provided by Korea Credit Guarantee Fund (KCGF) and the Korea Technology Credit Guarantee Fund (KOTEC)
DATES	Years when the programme was operating: 2001-2003 Evaluation period: 2000-2003 Year of the report: 2009 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to mitigate the negative effects of the financial crisis and to improve access to finance for SMEs.
	Yes	Improved access to finance, increased competitiveness, mitigation of negative effects of the financial crisis
EVALUATION THEME	Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	SMEs in Korea in manufacturing with at least five employees
SOURCE OF EVIDENCE	Academic article: (Oh et al., 2009[11]). Evaluation of credit guarantee policy using propensity score matching. Small Business Economics, 33(3), 335-351. Available at: https://doi.org/10.1007/s11187-008-9102-5
COUNTRY	Korea
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Total factor productivity (TFP), employment, sales, wage level, investment intensity, change in R&D status, firm survival
NON-SURVIVORS INCLUDED?	Yes, firm survival is an outcome variable.
DATA SOURCES	Administrative and survey: Annual Survey on Mining and Manufacturing in Korea in combination with data from the Korea Credit Guarantee Fund (KCGF) and Korea Technology Credit Guarantee Fund (KOTEC). 8 714 firms (100% of the population) supported by credit guarantees (3 996 firms were supported within KOTEC scheme, 3 818 firms within KCGF scheme and 900 firms within both schemes) supported between 2001-2002 and 35 299 non-supported firms.
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) in combination with a difference-in-differences (DiD) approach The authors estimate firm-level effects from t+1 to t+2
EVALUATION QUALITY SCORE	5
RELIABILITY COMMENTS	Analysis tackles the issue of industry bias and area bias. The authors analyse separately firms supported by KOTEC, KCGF and both schemes. The details about the control group concerning application for public support are missing.
KEY FINDINGS	The authors find positive effects on sales, employment, wage levels and survival rates. The remaining variables differed across both schemes. Firms supported by KOTEC scheme showed positive effects on changes in R&D status, and firms supported by both schemes showed positive effects on TFP. The remaining results were not conclusive.
PROGRAMME EXPENDITURE	The total amount of public resources allocated through both programmes in 2003 was nearly 12 trillion of Korean Won (1$ = 1037 KRW in November 2005).
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	No. The authors have not presented the results to the government.

Table B.11. The impact of investment support on labour productivity in Lithuanian family farms: A propensity score matching approach
TABLE REFERENCE	B11
PROGRAMME NAME	Modernisation of agricultural holdings, rural development programme
DATES	Years when the programme was operating: 2007-2012 Evaluation period: 2007-2012 Year of the report: 2019 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 1 Rural Development programme aimed to improve regional competitiveness of farms through the allocation of investment subsidies for modernisation of agricultural holdings.
	Yes	Higher competitiveness, higher investments
EVALUATION THEME	Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	Lithuanian family farms
SOURCE OF EVIDENCE	Academic article: (Namiotko et al., 2019[12]).The impact of Investment support on labour productivity in Lithuanian family farms: A propensity score matching approach. Economics and Sociology, 12(1), 342-352. Available at: https://doi.org/10.14254/2071-789X.2019/12-1/21
COUNTRY	Lithuania
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Labour productivity
NON-SURVIVORS INCLUDED?	No.
DATA SOURCES	Administrative data: The authors use Farm Accountancy Data Network (FADN) dataset obtained from Lithuanian Institute of Agrarian Economics. 284 farms were included in the empirical analysis. 62 farms were supported during 2007-2010 (5 445 farms were supported in total) and 222 non-supported (non-applicants).
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) in combination with a difference-in-differences (DiD) approach The authors estimate firm-level effects in t+2
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis does not tackle the issue of area bias, but it does address industry bias as it is a one-industry study. Firm size is not included in matching regression. We do not know the total number of supported recepients.
KEY FINDINGS	The authors conclude that Lithuanian farmers’ participation in investments promoting policy did not result in labour productivity gains.
PROGRAMME EXPENDITURE	Total financial allocation for the Rural Development programme was 2 524.7 mil. EUR. For the modernisation of agricultural holding was allocated 498.5 mill. EUR.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The results were not presented to the representatives of the government.

Table B.12. Mexico: Impact evaluation of SME programmes using panel firm data
TABLE REFERENCE	B12
PROGRAMME NAME	Entrepreneurship support programmes administered by various government agencies and ministries (Ministry of Economy, Nafinsa, Bancomext, Conacyt, Ministry of Labour). The evaluated programes include those run by: CIMO-PAC, FIDECAP, FAMPYME, Fondo PYME, COMPITE, CRECE, ROMODE, PROSEC, MEX-EX, PATCI, Crediexporta, PAT, PMT, PCI, PAIDEC, Fiscal Support and Technological Innovation
DATES	Years when the programme was operating: 2001-2006 Evaluation period: 1994-2005 Year of the report: 2010 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programmes differed in their stated objectives, but they seek to promote the productivity, quality, and competitiveness of small enterprises, encourage technology upgrading, training, and conservation, and improve earnings and safe working conditions for the workforce in firms.
	Yes	Increased competitiveness, increased hard and soft skills, innovation boost
EVALUATION THEME	Finance; Business advice, coaching, mentoring and counselling; Innovation
INTERVENTION TYPE	Both
TARGET GROUPS	All kinds of firms, depending on a specific programme
SOURCE OF EVIDENCE	Policy Research Working Paper: (Lopez-Acevedo and Tinajero, 2010[13]). Mexico: impact evaluation of SME programs using panel firm data. The World Bank, Available at: http://documents.worldbank.org/curated/en/421151468282531628/Mexico-impact-evaluation-of-sme-programs-using-panel-firm-data
COUNTRY	Mexico
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Employment, value added, gross production, sales, worked hours, wages, fixed assets, foreign sales, technology transfers payments, maquila services
NON-SURVIVORS INCLUDED?	No.
DATA SOURCES	Administrative and survey data: Data were obtained from surveys National Employment Salary, Training and Technology (Encuesta Nacional de Empleo, Salarios, Capacitación y Tecnología—ENESTYC) and Annual Industry Survey (Encuesta Industrial Annual—EIA) that are maintained by Mexico’s National Statistics Office (Instituto Nacional de Estadística y Geografía—INEGI) and from the programme data come from Ministry of Economy and National Science and Technology Council (Consejo Nacional de Ciencia y Tecnología—CONACyT). 838 firms supported (3 664 065 firms supported during 2001-2006 in total) and 1 540 non-supported (non-applicants).
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) with a difference-in-differences (DID) approach, fixed effects regressions The authors estimate mainly firm-level effects in t+2 (DID), but they also attempt to study long-term effects from t+1 to t+9 (and longer)
EVALUATION QUALITY SCORE	3
RELIABILITY COMMENTS	Analysis tackles the issue of area and selection bias. Nevertheless, the data on participation were collected retrospectively in 2001 and 2005 via a survey and thus they may be unreliable. The proportion of the sample analysed in the study is very small when compared with the population of treated firms. Different types of support (trainings, grants, financial instruments, technical assistance) are all mixed together in the findings.
KEY FINDINGS	The authors find that participation in any programme had a positive effect on value added, exports, sales, employment and fixed assets. However, the outcomes differed across the programmes. Once the authors separated the results across programmes, the results become very mixed, some of the schemes reported negative and non-significant results. The results across programmes are reported in the study.
PROGRAMME EXPENDITURE	Programmes administered by Ministry of Economy: 782 mil. USD Programmes administered by Nafinsa: 43,412 mil. USD Programmes administered by Bancomext: 34,449 mil. USD Programmes administered by Conacyt: 977 mil. USD Programmes administered by Ministry of Labor: 75 mil. USD The authors provide summary tables indicated number of recepients benefited from selected programmes and the total financial allocation.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The authors published two reports based on the study and they presented them to government officials. The focus of presentations was on how to better design the forthcoming policies.

Table B.13. Impotence of crisis-motivated subsidization of firms: The case of Slovenia
TABLE REFERENCE	B13
PROGRAMME NAME	Slovenia’s anti-crisis state aid programmes
DATES	Years when the programme was operating: 2009-2015 Evaluation period: 1998-2015 Year of the report: 2018 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programmes aimed to mitigate the negative effects of the financial crisis. SMEs received a financial subsidy for R&D activities, employment, training and rescuing and restructuring.
	Yes	Improved access to finance, mitigation of negative effects of the financial crisis, increased firm-survival
EVALUATION THEME	Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	SMEs in Slovenia
SOURCE OF EVIDENCE	Academic article: (Burger and Rojec, 2018[14]).Impotence of crisis-motivated subsidization of firms: The case of Slovenia. Eastern European Economics, 56(2), 122-148. Available at: https://doi.org/10.1080/00128775.2017.1416294.
COUNTRY	Slovenia
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Employment, sales
NON-SURVIVORS INCLUDED?	No.
DATA SOURCES	Administrative data: Firm-level data obtained from the Agency of the Republic of Slovenia for Public Legal Records and Related Services (AJPES) and from the Bank of Slovenia. Programme data were obtained from the Ministry of Finance State Aid. 24 385 firms supported by subsidies and 709 914 non-supported firms (non-applicants).
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) in combination with a difference-in-differences (DiD) approach The authors estimate firm-level effects from t+1 to t+5
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of industry bias and area bias. The analysis provides results for a long-term effects of public anti-crisis subsidies. Separate estimates across a type of subsidy are reported. Large-sample study.
KEY FINDINGS	The authors find that the programme increased employment, but it did not lead to an increase in sales.
PROGRAMME EXPENDITURE	The total amount of public resources allocated through the programmes was 688 mil. EUR.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The authors published results in a local language journal and a policy report, but the results were not presented to the government. According to authors judgement, the evaluation had no impact.

Table B.14. Loan guarantee schemes in the UK: the natural experiment of the enterprise finance guarantee and the 5 year rule
TABLE REFERENCE	B14
PROGRAMME NAME	Enterprise Finance Guarantee Scheme (EFG)
DATES	Years when the programme was operating: 2009-2013 (ongoing) Evaluation period: 2009-2013 Year of the report: 2018 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to facilitate access to finance through the allocation of loan guarantees in cooperation with commercial banks in response to distortions on the financial markets.
	Yes	Improved access to finance
EVALUATION THEME	Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	SMEs in the UK
SOURCE OF EVIDENCE	Academic article: (Cowling et al., 2018[15]). Loan guarantee schemes in the UK: the natural experiment of the enterprise finance guarantee and the 5 year rule. Applied Economics, 50(20), 2210-2218. Available at: https://doi.org/10.1080/00036846.2017.1392004
COUNTRY	United Kingdom
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Employment, sales
NON-SURVIVORS INCLUDED?	No.
DATA SOURCES	Survey data: Data provided by the UK Department for Business Innovation and Skills Enterprise Finance Guarantee (EFG). 500 firms supported by credit guarantees (6 504 supported in total in 2009), authors divided them into two groups: SMEs that would be supported only under Five Year Rule conditions (treated) and vs the remaining supported firms (control).
STEP LEVEL	6
METHODS	A natural experiment investigating whether a policy shift led to better economic firm-level outcomes or not Panel data approach OLS regressions The authors estimate firm-level effects from t+3
EVALUATION QUALITY SCORE	3
RELIABILITY COMMENTS	Analysis tackles the issue of industry bias but not area bias. Only t+3 effects are calculated, and the results are based on survey data.
KEY FINDINGS	The authors find positive effects on employment but they do not find positive effects on sales. The evaluation study indicates that a shift in a policy from a 5-Year Rule towards more relaxed conditions for applicants (higher credit guarantees and less constraints in terms of sales) for credit guarantee loans is less beneficial in terms of firm-level economic effects.
PROGRAMME EXPENDITURE	The supported enterprises could obtain a maximum loan guarantee of one mill. GBP. 23 762 loans with a total value of 2 106.7 mil. GBP were guaranteed between 2009-2013.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	N/A

Table B.15. Finance and growth at the firm level: evidence from SBA loans
TABLE REFERENCE	B15
PROGRAMME NAME	Small Business Administration (SBA) loans (lending programmes 7a and 504)
DATES	Years when the programme was operating: 1992-2007 Evaluation period: 1987-2012 Year of the report: 2017 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to increase employment through the allocation of subsidised loans.
	Yes	Improved access to finance, increase in employment
EVALUATION THEME	Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	Firms in the United States, focus on all kinds of firms
SOURCE OF EVIDENCE	Academic article: (Brown and Earle, 2017[16]). Finance and growth at the firm level: evidence from SBA loans. The Journal of Finance, 72(3), 1039-1080. Available at: https://doi.org/10.1111/jofi.12492
COUNTRY	United States
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Employment
NON-SURVIVORS INCLUDED?	No.
DATA SOURCES	Administrative data: Data for both groups were obtained from the Census Bureau. 128 900 firms were supported within scheme 7(a) and 28 600 firms were supported within the scheme 504. The control group was constructed from the population of non-recepients (500 000 firms).
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) in combination with a difference-in-differences (DiD) approach, instrumental variables (IV) approach, ordinary least squares (OLS) regressions The authors estimate firm-level effects from t+1 to t+5
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of industry bias and area bias. Different kinds of methods are used in to obtain more reliable results. Large sample study.
KEY FINDINGS	The authors find positive outcomes of the programme on employment. The authors report larger effects for younger and larger firms. The authors also report estimates on the effects across the intensity of public support.
PROGRAMME EXPENDITURE	N/A
MACRO IMPACT	The authors find positive outcomes of the programme on employment. The authors report larger effects for younger and larger firms. The authors also report estimates on the effects across the intensity of public support.
POLICY IMPACT OF THE EVALUATION	The authors have presented results to the government and discussed the evaluation in the U.S. Senate. As a result of the discussion, a Senate bill was introduced to increase loans in manufacturing.

Business Advice, Coaching, Mentoring, and Counselling

Table B.16. Publicly funded business advisory services and entrepreneurial outcomes
TABLE REFERENCE	B16
PROGRAMME NAME	Investment Network Programme administered by the Innovation Synergy Center
DATES	Years when the programme was operating: 2007-2009 Evaluation period: 2006-2009 Year of the report: 2012 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The main goal of the Investment Network Programme was to pursue entrepreneurial outcomes (i.e. performance and financial resources) in growth- and investment-oriented SMEs through provision of advisory services and counselling.
	Yes	Increased competitiveness, improved access to additional funding
EVALUATION THEME	Business advice, coaching, mentoring and counselling
INTERVENTION TYPE	Soft
TARGET GROUPS	Growth-oriented SMEs located in Ontario region
SOURCE OF EVIDENCE	Academic article: (Cumming and Fischer, 2012[17]). Publicly funded business advisory services and entrepreneurial outcomes. Research Policy, 41(2), 467-481. Available at: https://doi.org/10.1016/j.respol.2011.09.004
COUNTRY	Canada
REGIONAL/LOCAL	One region study, Ontario region
PERFORMANCE METRICS	Sales, obtaining an angel equity investment, patents, formation of a strategic alliance
NON-SURVIVORS INCLUDED?	Yes, 17 supported firms that went out of business are included in the analysis.
DATA SOURCES	Administrative data: Data were provided by the Investment Network. 101 treated firms and 127 non-supported (firms that were in touch with the Innovation Synergy Center, but not applied for the programme).
STEP LEVEL	6
METHODS	Panel data approach Two-stage Heckman selection model, Tobit regressions, Instrumental variables (IV) approach The authors estimate firm-level effects in t+1
EVALUATION QUALITY SCORE	5
RELIABILITY COMMENTS	Analysis tackles the issue of area and selection bias. The analysis contains a wide range of control variables, including characteristics of the top management team.
KEY FINDINGS	The authors find a positive impact of the programme on sales, patents, obtaining an angel equity investment and on formation of a strategic alliance.
PROGRAMME EXPENDITURE	The programme costs were totalled at 662 360 USD. The authors calculated that financing raised per dollar of cost at only 0.10 USD. Given that, the authors consider the programme to be cost-efficient.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The authors have presented the results to the Ontario Government and proposed several changes to the scheme, however, these changes were in the end voted out.

Table B.17. Supplier development programmes and firm performance: Evidence from Chile
TABLE REFERENCE	B17
PROGRAMME NAME	Chile Supplier Development Programme (Programa de Desarrollo de Proveedores - PDP) administered by the economic development agency CORFO.
DATES	Years when the programme was operating: 1998-ongoing Evaluation period: 1998-2008 Year of the report: 2013 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to promote, improve and stabilise mutually beneficial, long-term commercial relationships between large buyer firms — potential exporters — and their SME suppliers to increase competitiveness through tax-deductible expenses from corporate income tax. Eligible projects included purchases of specialised services, management training, technical assistance, advice and technology transfer services.
	Yes	Increased competitiveness, more intense cooperation between SMEs and large companies
EVALUATION THEME	Business advice, coaching, mentoring and counselling; Enterprise skills and culture
INTERVENTION TYPE	Soft
TARGET GROUPS	SMEs in Chile, evaluation focused on all firms in the agribusiness sector
SOURCE OF EVIDENCE	Academic article: (Arráiz, Henríquez and Stucchi, 2013[18]). Supplier development programs and firm performance: evidence from Chile. Small Business Economics, 41(1), 277-293. Available at: https://doi.org/10.1007/s11187-012-9428-x
COUNTRY	Chile
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Firm sustainability (positive sales), sales, export orientation (exporting), employment and wages
NON-SURVIVORS INCLUDED?	No
DATA SOURCES	Administrative data: Firm-level data from the Chilean tax administration agency (Servicios de Impuestos Internos - SII) and programme data from CORFO agency. The effects were estimated separately for sponsoring firms (large firms) and SMEs. There were 1 811 supported SMEs and 6 347 non-applicant SMEs, and 92 sponsoring firms and 9 916 non-applicant large firms.
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) in combination with a difference-in-differences (DiD) approach The authors estimate firm-level effects from t+1 to t+3
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of industry bias and area bias. It is an interesting evaluation indicating programme outcomes across two different groups of beneficiaries.
KEY FINDINGS	The authors find that both groups of firms (SMEs and large sponsor firms) benefited from the programme´s coordination efforts. The results show that SMEs improved their sales, employment, wages and sustainability, while large firms increased their sales and export orientation.
PROGRAMME EXPENDITURE	The programme expenditures during the years 2005-2008 were about 42.3 mil. USD.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The authors have presented evaluation results to the government, but they are not aware of any specific resulting policy change.

Table B.18. Assessing the effectiveness of guided preparation for new venture creation and performance: Theory and practice
TABLE REFERENCE	B18
PROGRAMME NAME	North Jutland Entrepreneurial Network (NiN) Programme
DATES	Years when the programme was operating: 2002-2006 Evaluation period: 2002-2008 Year of the report: 2012 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The main goal of the NiN programme is to guide and assist individuals engaged in the creation of a new venture as their primary occupation through the allocation of soft business support such as counselling and advisory. The programme offered three levels of counselling, including counselling provided by the local business centre (I), counselling with private-sector advisors (II), and extended counselling during the start-up with private-sector start-up consultants (III).
	Yes	Increase in start-ups, higher survival rates of new businesses
EVALUATION THEME	Business advice, coaching, mentoring and counselling
INTERVENTION TYPE	Soft
TARGET GROUPS	All SMEs located in North Jutland County except entrepreneurs doing business in agriculture, fisheries, fur and forestry
SOURCE OF EVIDENCE	Academic article: (Rotger, Gørtz and Storey, 2012[19]). Assessing the effectiveness of guided preparation for new venture creation and performance: Theory and practice. Journal of Business Venturing, 27(4), 506-521. Available at: https://doi.org/10.1016/j.jbusvent.2012.01.003
COUNTRY	Denmark
REGIONAL/LOCAL	One region study, North Jutland County, Denmark
PERFORMANCE METRICS	Survival, employment, numbers of firms with 20% growth in employment or sales
NON-SURVIVORS INCLUDED?	Yes, firm-survival is the main outcome variable.
DATA SOURCES	Administrative and survey data: The authors used programme data that were combined with administrative data from the Statistical Office of Denmark. Financial data available until 2006, survival data available until 2008. The authors worked with three treated groups of firms supported during 2002-2005. They divided them according to the intensity of support received. The control group was firms receiving a lower level of support. The numbers of treated firms are reported as follows: 932 enterprises supported by Level I (1 124 enterprises supported in total); 1 165 enterprises supported by Level II (1 541 enterprises supported in total), and 1 072 enterprises supported by Level III (1,525 enterprises supported in total).
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) with a difference-in-differences approach The authors estimate firm-level effects from t+1 to t+3
EVALUATION QUALITY SCORE	5
RELIABILITY COMMENTS	Analysis tackles the issue of area and selection bias. However, no group of non-applicant firms is used as an additional control group. Supported firms in cohorts 2002-2003, and 2004-2005 are analysed separately.
KEY FINDINGS	The authors find a positive impact of the programme on firm survival. The authors also report mostly positive effects on employment, turnover and growth. However, some of the coefficients were statistically insignificant.
PROGRAMME EXPENDITURE	The total amount of public resources allocated through the programmes was approximately 1 mil. USD. in 2009.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The authors have presented the results to policymakers. However, they are not aware of any specific policy changes driven by their evaluation.

Table B.19. The effect of business coaching on New Technology Based Firms: Survival–findings and lessons learned from a randomized controlled trial
TABLE REFERENCE	B19
PROGRAMME NAME	Business coaching programme for new technology-based firms
DATES	Years when the programme was operating: 2016-2017 Evaluation period: 2016-2017 Year of the report: 2019 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to increase survival rates of new technology-based firms through the facilitation of tactical business coaching.
	Yes	Increased firm survival, increase in entrepreneurial skills
EVALUATION THEME	Business advice, coaching, mentoring and counselling
INTERVENTION TYPE	Soft
TARGET GROUPS	New technology-based firms
SOURCE OF EVIDENCE	Academic article published in conference proceedings: (Ungerer et al., 2019[20]). The Effect of Business Coaching on NTBF Survival–Findings and Lessons Learned from a Randomized Controlled Trial. In Pallot, A., Zarli, A., Razek, A., R., A., Lecossier, A. (Eds.). 2019 IEEE International Conference on Engineering, Technology and Innovation (ICE/ITMC), 1-10. Available at: https://doi.org/10.1109/ICE.2019.8792604
COUNTRY	Germany
REGIONAL/LOCAL	One region study, experiment conducted in Baden-Wuerttemberg state
PERFORMANCE METRICS	Firm survival, business scale-up
NON-SURVIVORS INCLUDED?	Yes, this is the main outcome variable
DATA SOURCES	Survey data: Data were collected from a survey among participants. 36 treated firms (101 supported in total) and 57 non-supported firms.
STEP LEVEL	6
METHODS	Randomized controlled trial (RCT) The research team initially allocated firms randomly into treatment and control groups. However, after the experiment started, they did not manage to fulfil their initial selection and thus, the original samples were combined. The authors estimate effects in t+1 year
EVALUATION QUALITY SCORE	3
RELIABILITY COMMENTS	The analysis intended to tackle the issue of area and selection bias through conducting a randomized controlled trial. However, the study relies only on survey data and the authors fail in meeting implied analytical assumptions as they did not manage to control the firm distribution, i.e. to carry out the treatment to the firms according to the initial randomization into groups.
KEY FINDINGS	The authors do not find conclusive effects on firm survival. The authors provide a valuable lesson on the procedures of a randomized controlled trial.
PROGRAMME EXPENDITURE	The total amount of public resources allocated through both programmes was 500 mil. EUR.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The authors have presented the results to the Nesta, an organization that funded the RCT and that generally supports the execution of experiments in the field of business. The organisation may use the results to improve instructions for future studies, however, the authors are not familiar with the specific outcomes.

Table B.20. The impact of consulting services on small and medium enterprises: Evidence from a randomized trial in Mexico
TABLE REFERENCE	B20
PROGRAMME NAME	Business counselling services for SMEs in Puebla region
DATES	Years when the programme was operating: 2008-2009 Evaluation period: 2005-2014 Year of the report: 2018 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to expand the managerial skills of the owners/managers of SMEs by giving them access to subsidised consulting and mentoring services.
	Yes	Better firm performance, increase in entrepreneurial and managerial skills
EVALUATION THEME	Business advice, coaching, mentoring and counselling
INTERVENTION TYPE	Soft
TARGET GROUPS	SMEs
SOURCE OF EVIDENCE	Academic article: (Bruhn, Karlan and Schoar, 2018[21]). The impact of consulting services on small and medium enterprises: Evidence from a randomized trial in Mexico. Journal of Political Economy, 126(2), 635-687. Available at: https://doi.org/10.1086/696154
COUNTRY	Mexico
REGIONAL/LOCAL	One region study, experiment conducted in Puebla region
PERFORMANCE METRICS	Employment, total factor productivity, return on assets (ROA), wages, managerial and entrepreneurial skills
NON-SURVIVORS INCLUDED?	No
DATA SOURCES	Survey and administrative data: Data were obtained via follow-up survey and from administrative data on employment and wages from the Mexican Social Security Institute (IMSS). 150 treated firms and 282 non-supported firms.
STEP LEVEL	6
METHODS	Randomized controlled trial (RCT) Programme applicants were randomly allocated into treatment and control groups. The authors estimate effects from t+1 to t+5
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	The analysis tackles the issue of area and selection bias through conducting a randomized controlled trial (RCT). The RCT was conducted in a very transparent way.
KEY FINDINGS	The authors find positive effects on total factor productivity, return on assets, wages, employment and entrepreneurial skills.
PROGRAMME EXPENDITURE	The average cost of the consulting services was 11 856 USD per firm.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The authors presented the results to the government, but the unit that implemented the programme was later shut down for reasons unrelated to the programme, so they did not have a chance to continue or modify the programme.

Table B.21. Broader or deeper? Exploring the most effective intervention profile for public small business support
TABLE REFERENCE	B21
PROGRAMME NAME	Business Link Programme
DATES	Years when the programme was operating: 2003 Evaluation period: 2003-2005 Year of the report: 2011 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The Business Link Programme aimed to improve performance of small businesses in the UK through the allocation of soft business support, such as counselling and advisory services. The programme recipients received different types and intensities of soft support from the local Business Link Organisations and their partner organisations.
	Yes	Better firm performance, increase in skills of management and employees
EVALUATION THEME	Business advice, coaching, mentoring and counselling
INTERVENTION TYPE	Soft
TARGET GROUPS	SMEs in the UK
SOURCE OF EVIDENCE	Academic article: (Mole et al., 2011[22]). Broader or deeper? Exploring the most effective intervention profile for public small business support. Environment and Planning A, 43(1), 87-105. Available at:https://doi.org/10.1068/a43268
COUNTRY	United Kingdom
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Employment, sales, sales revenue per employee
NON-SURVIVORS INCLUDED?	No
DATA SOURCES	Survey and administrative data: Data were obtained via follow-up survey, from the Dun and Bradstreet UK database and from the government’s Small Business Service (SBS). A randomly selected sample of 2 296 supported firms and 1 152 non-supported firms in 2003. 1 130 firms received intensive support and 1 166 firms received less intensive support.
STEP LEVEL	6
METHODS	Two-stage Heckman selection model The authors estimate effects in t+1.5 (18 months)
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	The analysis tackles the issue of selection bias. However, it does not reflect the regional dimension of support recipients. The results are weighted across survey sampling. The authors report results across the intensity (less vs more intense) and type of public support (managed brokerage, light-touch brokerage, pipeline forcing, managed brokerage pipeline forcing).
KEY FINDINGS	The authors find a positive impact of the intensive support on employment and sales growth, and a negative impact of less intensive support on sales per employee.
PROGRAMME EXPENDITURE	The authors report that the average costs per company supported were 527.63 GBP.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The authors have presented results to the government and the evaluation was used to adjust the Business Link Programme in the shape of the new Growth Accelerator Programme.

Internationalisation

Table B.22. The effect of grant receipt on start-up size: Evidence from plant level data
TABLE REFERENCE	B22
PROGRAMME NAME	Grants for industrial development allocated by the Industrial Development Agency (IDA) and by Forbairt, Ireland
DATES	Years when the programme was operating: 1970-ongoing Evaluation period: 1972-2000 Year of the report: 2010 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to increase employment through investment subsidies (i.e. grants for industrial development) allocated to start-ups (new firms) in manufacturing.
	Yes	Increase in employment
EVALUATION THEME	Internationalisation; Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	New start-ups (plants) in manufacturing
SOURCE OF EVIDENCE	Academic article: (Girma et al., 2010[23]). The effect of grant receipt on start-up size: Evidence from plant level data. Journal of International Entrepreneurship, 8(4), 371-391. Available at: https://doi.org/10.1007/s10843-010-0061-y
COUNTRY	Ireland
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Employment
NON-SURVIVORS INCLUDED?	Yes, 10% of all start-ups had already closed down by the second year.
DATA SOURCES	Administrative and survey data: Data were provided by Forfás, the Irish policy and advisory board and by the Industrial Development Agency (IDA). 3 409 firms supported during 1972-2000 (of 3 901 firms supported in total during 1972-2000), 1 144 non-supported firms (non-applicants).
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) in combination with a difference-in-differences (DiD) approach, ordinary least squares (OLS) and quantile regressions The authors estimate firm-level effects in t+1
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of area and industry bias. Effects across firm size are reported.
KEY FINDINGS	The authors find positive effects of the programme on employment. They find greater effects for foreign firms when compared to domestic firms. They also found heterogeneous effects across the firm size.
PROGRAMME EXPENDITURE	The authors report that the average costs per company supported were 553 286 EUR.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	N/A

Innovation

Table B.23. Entrepreneurship policy and firm performance Chile’s CORFO seed capital programme
TABLE REFERENCE	B23
PROGRAMME NAME	Chile’s CORFO Seed Capital Programme
DATES	Years when the programme was operating: 2001-ongoing Evaluation period: 2008-2013 Year of the report: 2018 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme targets innovative, dynamic projects that would not otherwise be able to start up or grow. Beneficiaries receive a subsidy from CORFO agency to create and develop new innovative firms with high growth potential. The programme aims to boost start-ups’ sales and their ability to obtain external funding.
	Yes	Higher economic growth, increase in the number of high-growth start-ups
EVALUATION THEME	Innovation; Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	New high-growth oriented start-ups from Chile
SOURCE OF EVIDENCE	Academic article: (Navarro, 2018[24]). Entrepreneurship policy and firm performance Chile’s CORFO Seed Capital Program. Estudios de Economía, 45(2), 301-316. Available at: https://estudiosdeeconomia.uchile.cl/index.php/EDE/article/view/51345/53731
COUNTRY	Chile
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	New business formation, firm survival, increase in sales
NON-SURVIVORS INCLUDED?	Yes, firm survival is one of the outcome variables.
DATA SOURCES	Administrative data: Data were obtained from the CORFO agency. 376 projects supported during 2008-2012, 167 non-supported projects (rejected applicants). However, the authors acknowledge that initially there were 629 projects applying for the funding.
STEP LEVEL	5
METHODS	Panel data approach Ordinary least squares (OLS) regressions The authors estimate firm-level effects in t+1
EVALUATION QUALITY SCORE	2
RELIABILITY COMMENTS	The analysis does not fully tackle the issue of area and industry bias as the number of control variables is limited. A control group of non-applicants is not included in the analysis.
KEY FINDINGS	The authors find positive effects of the programme on the new business formation, firm survival and sales growth.
PROGRAMME EXPENDITURE	The average amount of public resources allocated to the recipients of the programme was 67 000 USD.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The author has presented results to the Chilean development agency and based on his recommendations, the policymakers focused more on the selection of the projects, when allocating public support.

Table B.24. Retaining winners: Can policy boost high-growth entrepreneurship?
TABLE REFERENCE	B24
PROGRAMME NAME	Finnish Governmental National Technology Agency’s (TEKES) programme NIY (Finnish acronym for young innovative growth companies)
DATES	Years when the programme was operating: 2008-2012 Evaluation period: 2006-2013 Year of the report: 2016 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to support young innovative and high-growth-oriented firms.
	Yes	Increase in number of fast-growing young ventures, increase in high-growth-oriented firms
EVALUATION THEME	Innovation; Business advice, coaching, mentoring and counselling
INTERVENTION TYPE	Hard & Soft
TARGET GROUPS	Young Finnish innovative ventures
SOURCE OF EVIDENCE	Academic article: (Autio and Rannikko, 2016[25]). Retaining winners: Can policy boost high-growth entrepreneurship?. Research policy, 45(1), 42-55. Available at:https://doi.org/10.1016/j.respol.2015.06.002
COUNTRY	Finland
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Sales
NON-SURVIVORS INCLUDED?	No
DATA SOURCES	Administrative and survey data: Data were obtained from National Technology Agencys (TEKES), from official statistics of financial records and other surveys. 56 firms supportedduring 2008-2010 in Sweden (of 160 firms supported in total), 101 non-supported Finnish firms (non-applicants) and (rejected applicants).
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) with a difference-in-differences approach The authors estimate firm-level effects from t+1 to t+3
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of area and selection bias.
KEY FINDINGS	The authors find that supported companies reported higher growth in sales by 120 percentage points compared to non-supported firms. Using multiplication analysis, the authors find that one Euro of public funding had generated 1.11 Euro of surplus sales growth (beyond trend growth) by 2013.
PROGRAMME EXPENDITURE	641.5 ths. EUR per supported venture (102.64 mil. EUR in total) allocated through the Finnish Government's National Technology Agency during 2008-2012 in Finland.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The results were presented to the government and disseminated to relevant stakeholders in Finland.

Table B.25. Can grants to consortia spur innovation and science-industry collaboration? Regression-discontinuity evidence from Poland
TABLE REFERENCE	B25
PROGRAMME NAME	Polish In-Tech programme on science-industry collaboration, research and innovation, and product commercialisation
DATES	Years when the programme was operating: 2012-2013 Evaluation period: 2012-2016 Year of the report: 2017 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The objective of the programme was to enhance the innovation activity of Polish enterprises through the allocation of innovation subsidies. The innovation subsidies were distributed by Poland’s National Centre for Research and Development (NCBiR) to partnership projects submitted by a team of a firm and a partner research institution (e. g. university).
	Yes	Increase in innovation activity
EVALUATION THEME	Innovation
INTERVENTION TYPE	Hard
TARGET GROUPS	Technology-oriented firms in Poland aiming to improve their innovation capacities
SOURCE OF EVIDENCE	Working paper: (Bruhn and McKenzie, 2019[26]). Can grants to consortia spur innovation and science-industry collaboration? Regression-discontinuity evidence from Poland. The World Bank Policy Research Working Series, Paper No. 7934. Available at: https://openknowledge.worldbank.org/bitstream/handle/10986/25943/WPS7934.pdf?sequence=1
COUNTRY	Poland
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Patent application, publication of a scientific article, citations, development of new industrial design, prototype, product, process, commercialisation of a new product/process, share of sales from new products/processes, new collaboration, commercialisation index, research and innovation index, collaboration index
NON-SURVIVORS INCLUDED?	No.
DATA SOURCES	Survey and administrative data: Programme data obtained from the National Center for Research and Development (NCBiR) and from the follow-up survey. 158 firms supported by the programme (164 firms were supported in total) and 301 non-supported firms (rejected applicants).
STEP LEVEL	6
METHODS	Panel data approach Regression discontinuity design (RDD) The authors estimate firm-level effects from t+2.5 to t+3.5 (years)
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	The analysis tackles the issue of area and industry bias. Most of the outcome variables are measured through a results of survey. Wide range of outcome variables.
KEY FINDINGS	The authors find that the programme improved science-industry collaboration, increased the probability of applying for a patent and probability of publishing an academic article, and they also report positive effects on the commercialisation of new products/processes (including sales).
PROGRAMME EXPENDITURE	The average amount of public resources allocated to the recipients of the programme was 660 000 USD.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The authors presented the results to the government, and report that the representatives of the public administration said that their evaluation contributed to encouragement of evidence-based innovation policy design in Poland. The evaluation was further used to support an extension of programmes focusing on innovation grants, R&D grants, and science and technology grants provided by NCBiR.

Table B.26. Do selected firms show higher performance? The case of Portugal’s innovation subsidy
TABLE REFERENCE	B26
PROGRAMME NAME	Portuguese Innovation Incentive System (SI Innovation), an instrument of the National Strategic Reference Framework (NSRF), included in the Operational Programme for Competitiveness Factors (COMPETE)
DATES	Years when the programme was operating: 2007-2013 Evaluation period: 2006-2016 Year of the report: 2019 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to promote innovation in the business sector, support firms’ progression in the value chain, their orientation to international markets, and stimulate qualified entrepreneurship and investments in new areas with growth potential.
	Yes	Increased innovation, increased competitiveness
EVALUATION THEME	Innovation
INTERVENTION TYPE	Hard
TARGET GROUPS	New and established firms with innovation potential operating in Portugal
SOURCE OF EVIDENCE	Academic article: (Santos, 2019[27]). Do selected firms show higher performance? The case of Portugals innovation subsidy. Structural Change and Economic Dynamics, 50, 39-50. Available at: https://doi.org/10.1016/j.strueco.2019.04.003
COUNTRY	Portugal
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Employment, sales, EBITDA, gross value added (GVA), labour productivity, total factor productivity (TFP), value creation, tangible fixed assets, patent stock
NON-SURVIVORS INCLUDED?	No
DATA SOURCES	Administrative and commercial data: Firm-level data from the Amadeus database (Bureau van Dijk), in combination with data from the Portuguese National Institute of Statistics (INE) and the Information System of the Portuguese NSRF Incentive Systems. 134 firms supported by soft loans (of about 2 600 firms supported in total) and 186 non-supported firms (rejected applicants).
STEP LEVEL	5
METHODS	Panel data approach Propensity score matching (PSM) in combination with a difference-in-differences (DiD) approach The authors estimate firm-level effects from t+1 to t+4
EVALUATION QUALITY SCORE	3
RELIABILITY COMMENTS	Analysis tackles the issue of industry bias and area bias. Long-term effects are analysed. Representation of treated firms is, however, rather small. An additional control group of non-applicants is missing.
KEY FINDINGS	The evaluation finds positive effects on investments, sales, technological progress and job creation, however, negative effects on labour productivity and value creation.
PROGRAMME EXPENDITURE	The total amount of public resources allocated through SI Innovation was 2 000 mil. EUR.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The author has sent the results to the government officials in charge of European Union funding.

Table B.27. The impact of government-supported participative loans on the growth of entrepreneurial ventures
TABLE REFERENCE	B27
PROGRAMME NAME	EBT and PYME participative loans (loan contracts) programmes allocated by the governmental agency Empresa Nacional de Innovacin (ENISA).
DATES	Years when the programme was operating: 2005-2011 Evaluation period: 2005-2014 Year of the report: 2019 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 Young ventures were supported by two programmes, the PYME programme that aimed to support high-growth entrepreneurial ventures and the EBT programme that supported high-technology firms.
	Yes	Improved access to finance, increased competitiveness
EVALUATION THEME	Innovation; Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	Young entrepreneurial SMEs in Spain
SOURCE OF EVIDENCE	Academic article: (Bertoni, Martí and Reverte, 2019[28]). The impact of government-supported participative loans on the growth of entrepreneurial ventures. Research Policy, 48(1), 371-384. Available at: https://doi.org/10.1016/j.respol.2018.09.006
COUNTRY	Spain
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Employment, sales, survival rate
NON-SURVIVORS INCLUDED?	Yes, it is one of the outcome variables.
DATA SOURCES	Administrative and accounting data:Accounting data (source not reported) and administrative data obtained from a governmental agency ENISA. 512 firms established after 2003 that received a participative loan from ENISA between 2005 and 2011 (of 293 firms supported by EBT and 466 firms supported by PYME in total) and a control group of 9 050 firms founded in Spain between 2003 and 2011 (randomly selected).
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) in combination with a difference-in-differences (DiD) approach and 2-stage least squares (2SLS), GMM, fixed effects and OLS regressions The authors estimate firm-level effects from t+1 to t+2
EVALUATION QUALITY SCORE	5
RELIABILITY COMMENTS	Analysis tackles the issue of industry bias and area bias. Many different evaluation methods (PSM + DID, 2SLS regressions, dynamic panel regressions) are used. Separate effects for young, small and high-tech firms, intensity of public support and during the financial crisis are reported. The authors analyse together the EBT and PYME schemes and programmes. Separate results across programmes are not reported.
KEY FINDINGS	The authors find positive effects on employment and sales. The effects are larger for high-tech, young and small entrepreneurial ventures and for those that received a participative loan during the global financial crisis. The evaluation failed to find a statistically significant effect on survival rates.
PROGRAMME EXPENDITURE	The total amount of public resources allocated through both programmes during 2005-2011 was 263.5 mil. EUR.
MACRO IMPACT	The authors estimate that one mil. EUR invested in participative loans generates 12 114.7 jobs and 1 091.97 mil. EUR in sales.
POLICY IMPACT OF THE EVALUATION	The evalution was presented to the ENISA institution and to other high-rank officials of the government. It served to endorse the evaluated schemes as well as a new scheme for young entrepreneurs that started as a pilot programme in 2010.

Table B.28. Inside the black box of outcome additionality: Effects of early-stage government subsidies on resource accumulation and new venture performance
TABLE REFERENCE	B28
PROGRAMME NAME	VINN NU (Win Now) programme operated by the Swedish Governmental Agency for Innovation Systems (VINNOVA)
DATES	Years when the programme was operating: 2002-2008 Evaluation period: 2001-2011 Year of the report: 2015 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 3 The programme aimed to support new innovative and high-growth-oriented firms.
	Yes	Increase in number of new and innovative start-ups, increase in number of high-growth-oriented firms
EVALUATION THEME	Innovation; Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	New Swedish innovative ventures
SOURCE OF EVIDENCE	Academic article: (Söderblom et al., 2015[29]). Inside the black box of outcome additionality: Effects of early-stage government subsidies on resource accumulation and new venture performance. Research Policy, 44(8), 1501-1512. Available at:https://doi.org/10.1016/j.respol.2015.05.009
COUNTRY	Sweden
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Employment, equity, sales
NON-SURVIVORS INCLUDED?	Yes, 13% of the sample went out of business.
DATA SOURCES	Administrative data: Data were obtained from Swedish Governmental Agency for Innovation Systems (VINNOVA). 130 firms supportedduring 2002-2008 in Sweden (100% of population), 154 non-supported firms from Sweden (rejected applicants).
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) with a difference-in-differences approach The authors estimate firm-level effects from t+1 to t+7
EVALUATION QUALITY SCORE	5
RELIABILITY COMMENTS	Analysis tackles the issue of area and selection bias.
KEY FINDINGS	The authors find positive effects on employment, sales and external equity funding.
PROGRAMME EXPENDITURE	28 ths. EUR per supported venture (3.64 mil. EUR in total) allocated through Swedish Governmental Agency for Innovation Systems during 2002-2008 in Sweden.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The results from the study were presented on several occasions to government representatives and other policymakers, including Swedish Governmental Agency for Innovation Systems (VINNOVA). VINNOVA launched a new and more extensive programme, Innovative Startups, replacing the programme the study investigated, and the study had some impact on that decision.

Table B.29. Impact of Swiss technology policy on firm innovation performance: an evaluation based on a matching approach
TABLE REFERENCE	B29
PROGRAMME NAME	Swiss innovation policy administered by the Commission of Technology and Innovation (CTI)
DATES	Years when the programme was operating: 2000-2002 Evaluation period: 2000-2004 Year of the report: 2010 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The objective of the programme was to enhance the innovation activity of Swiss enterprises through the allocation of innovation subsidies. The innovation subsidies are distributed to partnership projects that have been submitted by a team of firm and a partner research institution (e. g. university).
	Yes	Increase in innovation activity
EVALUATION THEME	Innovation; Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	Technology-oriented firms in Switzerland aiming to improve their innovation capacities
SOURCE OF EVIDENCE	Academic article: (Arvanitis, Donzé and Sydow, 2010[30]). Impact of Swiss technology policy on firm innovation performance: an evaluation based on a matching approach. Science and Public Policy, 37(1), 63-78. Available at: https://doi.org/10.3152/030234210X491623
COUNTRY	Switzerland
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Share of sales from new products, share of sales from new markets worldwide, percentage increase in sales, percentage reduction of average variable production costs due to the innovation process, economic importance of the innovations introduced, technical importance of the innovations introduced
NON-SURVIVORS INCLUDED?	No.
DATA SOURCES	Survey and administrative data: Firm-level data obtained from the Commission of Technology and Innovation (CIT). 199 firms supported by the programme (307 firms were supported in total) and 996 non-supported firms (non-applicants) that participated in the Swiss Innovation Survey 2002 and reported the introduction of innovations in the period 2000-2002.
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) in combination with a difference-in-differences (DiD) approach The authors estimate firm-level effects from t+1 to t+2
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	The analysis tackles the issue of area and industry bias. Most of the outcome variables are measured through a results survey and two of them as ordinary variables.
KEY FINDINGS	The authors find that the programme improved the innovation performance of supported firms with respect to six different measures of innovation performance.
PROGRAMME EXPENDITURE	The total amount of public resources allocated the programme during the years 2000-2002 was 120 mil. Swiss francs (CHF).
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The results were presented to the Commission of Technology and Innovation (CTI) and the study encouraged usage of econometric methods for programme evaluation for projects financed by the CTI.

Table B.30. Evaluating effectiveness of public support to business R&D in Türkiye through concepts of input and output additionality
TABLE REFERENCE	B30
PROGRAMME NAME	The TUBITAK-TEYDEB Public R&D Programme administered by the Scientific and Technological Research Council of Türkiye (TUBITAK)
DATES	Years when the programme was operating: 1995-ongoing Evaluation period: 2003-2006 Year of the report: 2011 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The objective of the programme was to enhance the international competitiveness of industrial companies in Turkey by means of higher R&D and innovation expenditures.
	Yes	Increase in innovation activity, increased competitiveness
EVALUATION THEME	Innovation; Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	Technology-oriented firms in Türkiye aiming to improve their innovation capacities
SOURCE OF EVIDENCE	Academic working paper: (Tandogan and Pamukcu, 2011[31]). Evaluating effectiveness of public support to business R&D in Turkey through concepts of input and output additionality. Economic Research Forum Working Paper 593, The Economic Research Forum (ERF), Egypt. Available at: http://erf.org.eg/wp-content/uploads/2014/08/593.pdf
COUNTRY	Türkiye
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	R&D intensity, R&D expenditures per employee, share of R&D personnel, export intensity, import intensity
NON-SURVIVORS INCLUDED?	No.
DATA SOURCES	Survey and administrative data: Firm-level data from the Structural Business Statistics Survey (SBS), Foreign Trade Statistics, General Census of Industry and Establishments and Producers Price Index collected by the Turkish Statistical Institute (Turkstat) and administrative data maintained by the governmental agency TUBITAK. 97 firms supported by the programme in 2004 (326 firms were supported in 2004 in total) and 6 511 non-supported firms (non-applicants).
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) in combination with a difference-in-differences (DiD) approach The authors estimate firm-level effects from t+1 to t+2
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	The analysis does not tackle the issue of area bias. It addresses only industry bias. The authors do not include regional variables in the matching regression.
KEY FINDINGS	The authors find only a positive change in share of R&D personnel, R&D expenditures per employee and R&D intensity for the programme participants. The effects for the remaining variables were not found to be statistically significant.
PROGRAMME EXPENDITURE	The total amount of public resources allocated the programme in 2004 was 491 mil. USD.
MACRO IMPACT	The authors follow-up on time series of the R&D outcome indicators for the whole country and they assume a positive trend related to the public programme, however, no direct associations are tested.
POLICY IMPACT OF THE EVALUATION	The authors presented the study to the policymakers and stakeholders, however, they are not aware of any specific changes implemented based on their evaluation.

Table B.31. Boon or boondoggle? Business incubation as entrepreneurship policy
TABLE REFERENCE	B31
PROGRAMME NAME	Business incubators in the United States.
DATES	Years when the programme was operating: 1990-2007 Evaluation period: 1990-2008 Year of the report: 2010 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to support new innovative and high-growth-oriented firms (depending on the type of incubator).
	Yes	Employment growth, increase in the firms survival rates, increase in number of high-growth-oriented firms
EVALUATION THEME	Innovation; Finance
INTERVENTION TYPE	Hard & Soft
TARGET GROUPS	New businesses less than 5 years old
SOURCE OF EVIDENCE	Published doctoral dissertation (Academic): (Amezcua, 2010[32]).Boon or Boondoggle? Business Incubation as Entrepreneurship Policy,ProQuest Dissertations Publishing,Syracuse University. Available at:https://search.proquest.com/docview/874370586?accountid=17203.
COUNTRY	United States
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Survival, sales, employment
NON-SURVIVORS INCLUDED?	Yes, survival is one of the outcome variables.
DATA SOURCES	Administrative and survey data: Data were obtained from a panel of demographic information on business incubators from the National Business Incubation Association, and from the National Establishment Time-Series Database (NETS). 18 426 firms incubated firms (from 65 incubators), 28 346 non-incubated firms.
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) with a difference-in-differences approach,generalized method of moments (GMM) regressions,Hausman-Taylor generalized IV regressions The author estimates firm-level effects from t+1
EVALUATION QUALITY SCORE	5
RELIABILITY COMMENTS	Analysis tackles the issue of area and selection bias.
KEY FINDINGS	The results show that incubated firms have a slightly lower survival rate, but slightly higher employment and sales growth than non-incubated firms. The evaluation finds that firms from university-sponsored incubators report better results compared to other incubators. The results also show that tenants of profit-oriented incubators outperform firms from non-profit incubators.
PROGRAMME EXPENDITURE	N/A
MACRO IMPACT	The author predicted annual growths of survival, sales and employment for the population of incubated and non-incubated firms. The prediction shows, based on employment and sales performance, that incubation generally has a positive effect but there are net losses in employment and sales for the incubated group. Firms in incubation are better off than had they not been incubated, but they are still more likely to fail and not grow.
POLICY IMPACT OF THE EVALUATION	N/A

Enterprise Skills and Culture

Table B.32. Counterfactual impact evaluation on EU cohesion policy interventions in training in companies
TABLE REFERENCE	B32
PROGRAMME NAME	The Human Resources and Employment Operational Programme (HREOP), Czech Republic
DATES	Years when the programme was operating: 2007-2013 Evaluation period: 2008-2012 Year of the report: 2016 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to support the competitiveness of companies through the development of professional knowledge, competence and improvement in the qualification of employees. Supported companies benefitted from various training activities. These were especially focused on modern management methods and human resource management.
	Yes	Higher firm performance, increase in employee and management skills
EVALUATION THEME	Enterprise skills and culture
INTERVENTION TYPE	Soft
TARGET GROUPS	Firms in the Czech Republic
SOURCE OF EVIDENCE	Academic article: (Potluka et al., 2016[33]). Counterfactual Impact Evaluation on EU Cohesion Policy Interventions in Training in Companies. Ekonomicky Casopis, 64(6), 575-595. Available at: https://www.ceeol.com/search/article-detail?id=443303
COUNTRY	Czech Republic
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Employment
NON-SURVIVORS INCLUDED?	No.
DATA SOURCES	Administrative data: Data were collected from the programme monitoring system Monit7+ and from the Czech Statistical Office (CZSO). 373 treated firms supported during 2009-2012 (of 1 447 firms supported in total), 202 non-supported firms (of 1 183 rejected applicants in total).
STEP LEVEL	6
METHODS	Panel data approach Instrumental variables (IV) approach The authors estimate firm-level effects in t+1
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of area and selection bias.
KEY FINDINGS	The authors do not find statistically significant positive effects on employment.
PROGRAMME EXPENDITURE	The total amount of public resources allocated through trainings to the firms was 618 mil. EUR.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The results were presented at the Ministry of Labour and Social Affairs for the relevant stakeholders. The authors recommended to the Ministry representatives to focus training support on smaller firms and only on hard skills. These recommendations were adopted by the Ministry in the subsequent programming period. Specifically training activities focusing on soft skills have been removed from the eligible list of training activities.

Table B.33. The impact of entrepreneurship education on entrepreneurship skills and motivation
TABLE REFERENCE	B33
PROGRAMME NAME	Junior Achievement Young Enterprise student mini-company (SMC) programme coordinated by the Jong Ondernemen Association
DATES	Years when the programme was operating: 2005 Evaluation period: 2005-2006 Year of the report: 2010 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to improve (build) entrepreneurial competences, skills and intentions of young students through the student mini-company (SMC) programme. The students were involved in the development of a small-sized and short-duration business from its early establishment supported by one or two advisers from the business world.
	Yes	Higher entrepreneurial competences, skills and intentions of students
EVALUATION THEME	Enterprise skills and culture
INTERVENTION TYPE	Soft
TARGET GROUPS	University students in areas of management, economics and law
SOURCE OF EVIDENCE	Academic article: (Oosterbeek, van Praag and Ijsselstein, 2010[34]). The impact of entrepreneurship education on entrepreneurship skills and motivation. European Economic Review, 54(3), 442-454. Available at: https://doi.org/10.1016/j.euroecorev.2009.08.002.
COUNTRY	Netherlands (However, the programme operates worldwide)
REGIONAL/LOCAL	One country study, the evaluation focused on the three locations of one University (AVANS Hogeschool)
PERFORMANCE METRICS	Entrepreneurial competences and intentions (validated scales) measured as need for achievement, need for autonomy, need for power, social orientation, self efficacy, endurance, risk taking propensity, market awareness, creativity, flexibility
NON-SURVIVORS INCLUDED?	No.
DATA SOURCES	Survey data: Data were collected via non-anonymous surveys at the presences of staff and lecturers. 104 students that participated in the programme (189 students participated in total), 146 non-supported students (non-participants).
STEP LEVEL	5
METHODS	Panel data approach A difference-in-differences (DID) approach and a DID combined with an instrumental variables (IV) approach The authors estimate firm-level effects in t+1
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of area and selection bias. However, the control variables do not include student´s home town´s (regional) variables. Effects across gender are reported.
KEY FINDINGS	The authors find that students’ self-assessed entrepreneurial skills (and traits) did not improve after the participation in the programme and the effect of the programme on entrepreneurial intentions was found to be negative.
PROGRAMME EXPENDITURE	N/A
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	N/A

Table B.34. The effect of a tax training programme on tax compliance and business outcomes of starting entrepreneurs: Evidence from a field experiment
TABLE REFERENCE	B34
PROGRAMME NAME	Netherlands' Tax and Customs Administration (NTCA) tax training programme
DATES	Years when the programme was operating: 2008-2009 Evaluation period: 2008-2012 Year of the report: 2019 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The training programme aimed to increase tax compliance of newly established entrepreneurs and to improve their business outcomes as a result of more efficient dealing with tax authorities and business legislation.
	Yes	Increased tax compliance, higher firm performance
EVALUATION THEME	Enterprise culture and skills
INTERVENTION TYPE	Soft
TARGET GROUPS	New first-time entrepreneurs
SOURCE OF EVIDENCE	Academic article: (Nagel et al., 2019[35]). The effect of a tax training program on tax compliance and business outcomes of starting entrepreneurs: Evidence from a field experiment. Journal of Business Venturing, 34(2), 261-283. Available at: https://doi.org/10.1016/j.jbusvent.2018.10.006
COUNTRY	Netherlands
REGIONAL/LOCAL	One country study, the programme took place in East Netherlands
PERFORMANCE METRICS	Firm survival, profit, business costs, filing tax return correct, complete and in time, and paying the amount of taxes due in time, bookkeeping skills
NON-SURVIVORS INCLUDED?	Yes, it is one of the outcome variables.
DATA SOURCES	Survey and administrative data: Data were collected via surveys and from the Netherlands' Tax and Customs Administration. 352 new entrepreneurs that participated in the programme, 466 non-supported entrepreneurs (non-participants).
STEP LEVEL	6
METHODS	Randomized controlled trial (RCT) Programme applicants were randomly allocated into treatment and control groups. The authors estimate effects from t+1 to t+3
EVALUATION QUALITY SCORE	5
RELIABILITY COMMENTS	Analysis tackles the issue of area and selection bias through conducting a randomized controlled trial (RCT). The RCT was conducted in a very transparent way.
KEY FINDINGS	The authors find positive effects on profit due to different handling of business costs and some areas of tax compliant behaviour, however, they find no impact on firm survival.
PROGRAMME EXPENDITURE	N/A
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The study examined one of the changes within the NTCA’s strategy and was presented to and discussed with the NTCA. The results of the study led to continuation of the training programme in an adapted version.

Table B.35. The impact of employees' and managers' training on the performance of small‐and medium‐sized enterprises: Evidence from a randomized natural experiment in the UK service sector
TABLE REFERENCE	B35
PROGRAMME NAME	Business Innovation and Skills (BIS) training programme
DATES	Years when the programme was operating: 2002-2003 Evaluation period: 2002-2006 Year of the report: 2016 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to expand the skills of the employees and the general management and human resource management skills of owners/managers of SMEs, through the allocation of business training activities (i.e. employee training, human resource management training, and manager training).
	Yes	Higher firm performance, increase in skills of management and employees
EVALUATION THEME	Enterprise skills and culture; Business advice, coaching, mentoring and counselling
INTERVENTION TYPE	Soft
TARGET GROUPS	SMEs in the UK service sector
SOURCE OF EVIDENCE	Academic article: (Georgiadis and Pitelis, 2016[36]). The impact of employees' and managers' training on the performance of small-and medium-sized enterprises: Evidence from a randomized natural experiment in the UK service sector. British Journal of Industrial Relations, 54(2), 409-421. Available at:https://doi.org/10.1111/bjir.12094
COUNTRY	United Kingdom
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Profit margin, sales revenue per employee
NON-SURVIVORS INCLUDED?	No
DATA SOURCES	Survey data: Data were obtained via follow-up survey. The final sample consisted of 430 firms responding to the survey, out of which 287 firms received at least one training activity (845 firms received at least one training activity in total) and 143 represented the control group (the initial size of the control group was 480 firms).
STEP LEVEL	6
METHODS	Randomized controlled trial (RCT) Programme applicants were randomly allocated into treatment and control groups. The authors estimate effects in t+2
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of area and selection bias through conducting a randomized controlled trial (RCT). The analysis relies on survey data.
KEY FINDINGS	The authors generally find positive effects on profit margin and sales revenue per employee for firms participating in at least one training activity. However, the authors also study the effects of different training activities, and there the effects vary across the type of training.
PROGRAMME EXPENDITURE	N/A
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The authors are not aware of any policy impact of the evaluation.

Table B.36. Behind the GATE experiment: Evidence on effects of and rationales for subsidized entrepreneurship training
TABLE REFERENCE	B36
PROGRAMME NAME	Project Growing America through Entrepreneurship (GATE)
DATES	Years when the programme was operating: 2003-2005 Evaluation period: 2003-2005 Year of the report: 2015 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 GATEs objective was to help emerging entrepreneurs in rural and urban communities to establish their own business. The programme randomly offered free entrepreneurship training to individuals (applicants) interested in starting or improving their established business.
	Yes	Increase in start-ups, increase in entrepreneurial skills
EVALUATION THEME	Enterprise skills and culture
INTERVENTION TYPE	Soft
TARGET GROUPS	Individuals interested in starting or improving their business, focused mainly on unemployed individuals
SOURCE OF EVIDENCE	Academic article: (Fairlie, Karlan and Zinman, 2015[37]). Behind the GATE experiment: Evidence on effects of and rationales for subsidized entrepreneurship training. American Economic Journal: Economic Policy, 7(2), 125-61. Available at:https://doi.org/10.1257/pol.20120337
COUNTRY	United States
REGIONAL/LOCAL	One country study, training is offered at seven sites in three states
PERFORMANCE METRICS	Business start-up, household income, employment, sales
NON-SURVIVORS INCLUDED?	Yes, 20% firms went out of business.
DATA SOURCES	Administrative data and survey data: Data were obtained from the US Department of Labor and the Small Business Administration (SBA) and from a survey among participants 2 094 participants and 2 103 non-participants (applicants)
STEP LEVEL	6
METHODS	Randomized controlled trial (RCT) Programme coordinators randomized applicants to treatment or control with equal probability The authors estimate effects (Difference-in-Differences DID) in t+6, t+18 and t+60 months
EVALUATION QUALITY SCORE	5
RELIABILITY COMMENTS	Analysis tackles the issue of area and selection bias.
KEY FINDINGS	The authors find positive short-term effects on business start-up, but the effects mitigated in the long-run. The authors do not find positive effects on business performance.
PROGRAMME EXPENDITURE	Total costs of providing training to GATE recipients were estimated as 1 321 USD per person (approximately 2 766 ths. USD). The authors conclude that the programme was not cost-efficient.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The results were presented to the representatives of the U.S. Small Business Administration. Because of the findings, the U.S. Department of Labor asked for help in designing a new pilot programme, which included some of the recommendations the authors made. In particular, it provides a small amount of seed capital and targets only individuals with previous experience in the same industry as the proposed business. The pilot programme is under evaluation.

Inclusive Entrepreneurship

Table B.37. The effects of micro-entrepreneurship programmes on labour market performance: experimental evidence from Chile
TABLE REFERENCE	B37
PROGRAMME NAME	Micro-entrepreneurship Support Programme (MESP)
DATES	Years when the programme was operating: 2002-ongoing Evaluation period: 2010-2014 Year of the report: 2018 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to provide individuals with the skills and capital required to generate income through self-employment by developing their own businesses. The programme targets individuals from extremely poor households who receive start-up subsidy, training and mentoring.
	Yes	Poverty reduction, increase in earnings, increase in start-ups
EVALUATION THEME	Inclusive entrepreneurship; Business advice, coaching, mentoring and counselling; Finance
INTERVENTION TYPE	Hard & Soft
TARGET GROUPS	Poor individuals (i.e. beneficiaries of the anti-poverty programme)
SOURCE OF EVIDENCE	Academic article: (Martínez, Puentes and Ruiz-Tagle, 2018[38]). The effects of micro-entrepreneurship programs on labor market performance: experimental evidence from Chile. American Economic Journal: Applied Economics, 10(2), 101-24. Available at: https://doi.org/10.1086/696154
COUNTRY	Chile
REGIONAL/LOCAL	One region study, experiment conducted in Santiago Metropolitan Area.
PERFORMANCE METRICS	Employment, earnings
NON-SURVIVORS INCLUDED?	No
DATA SOURCES	Survey and administrative data: Data were obtained via follow-up survey and from the administrative data on employment and wages from the Unemployment Insurance (UI) system. There were 689 treated individuals in the MESP programme, and 693 idividuals in the MESP+ programme (receiving an extra subsidy) and 566 individuals in the control group.
STEP LEVEL	6
METHODS	Randomized controlled trial (RCT) Programme applicants were randomly allocated into treatment and control groups. The authors estimate effects from t+2 to t+3
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of area and selection bias through conducting a randomized controlled trial (RCT). The RCT was conducted in a very transparent way. Separate results for MESP and MESP+ programme participants are reported.
KEY FINDINGS	The authors find positive effects on employment and earnings, however, the effects are decreasing over time in the long run.
PROGRAMME EXPENDITURE	The authors conduct cost-benefit analysis of the programme by comparing the total labor income increase to the programme’s direct costs which are 1 200 USD for MESP participants and 1 440 USD for MESP+ participants (1.83 mil. USD in total). They conclude that the programme´s direct benefits exceed the costs.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	Based on the evaluation efforts, the programme has been in operation over time until the present day without significant modification. The programme has been continuously evaluated on an annual basis.

Table B.38. Long term effect of public subsidies on start-up survival and economic performance: An empirical study with French data
TABLE REFERENCE	B38
PROGRAMME NAME	ACCRE start-up support for the unemployed (Aide aux chômeurs créant ou reprenant une entreprise)
DATES	Years when the programme was operating: 1998 Evaluation period: 1998-2006 Year of the report: 2015 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to activate unemployed individuals through public support (lump-sum payment) and encourage them to become self-employed.
	Yes	Reduction of unemployment, increase in start-ups
EVALUATION THEME	Inclusive entrepreneurship; Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	Unemployed individuals
SOURCE OF EVIDENCE	Academic article: (Duhautois, Redor and Desiage, 2015[39]). Long Term Effect of Public Subsidies on Start-up Survival and Economic Performance: An Empirical Study with French Data?. Revue d'conomie industrielle, 149(1), 11-41. Available at:https://journals.openedition.org/rei/6063
COUNTRY	France
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Survival
NON-SURVIVORS INCLUDED?	Yes, firm survival is the main outcome variable.
DATA SOURCES	Administrative and survey data: Data were obtained from the SINE survey, INSEE (the French Institute of Statistics) and from an administrative database FICUS. 1 960 entrepreneurs supported in 1998 in France and 2 643 non-supported entrepreneurs (non-applicants).
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) with a difference-in-differences approach The authors estimate firm-level effects from t+1 to t+8
EVALUATION QUALITY SCORE	5
RELIABILITY COMMENTS	Analysis tackles the issue of area and selection bias. The authors combine entrepreneur and firm-level data and they study long-term effects of the programme.
KEY FINDINGS	The authors find a positive long-term impact of the programme on firm survival.
PROGRAMME EXPENDITURE	The total amount of public resources allocated through the programme was 700 mil. EUR.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	No. The authors have not presented the results to the government.

Table B.39. You can go your own way! The long‐term effectiveness of a self‐employment programme for welfare recipients in Germany
TABLE REFERENCE	B39
PROGRAMME NAME	German start-up subsidy programme Einstiegsgeld
DATES	Years when the programme was operating: 2005-ongoing Evaluation period: 2005-2011 Year of the report: 2016 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to activate unemployed individuals through start-up subsidy and encourage them to become self-employed.
	Yes	Reduction of unemployment, increase in start-ups
EVALUATION THEME	Inclusive entrepreneurship; Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	Unemployed individuals
SOURCE OF EVIDENCE	Academic article: (Wolff, Nivorozhkin and Bernhard, 2016[40]). You can go your own way! The long-term effectiveness of a self-employment programme for welfare recipients in Germany. International Journal of Social Welfare, 25(2), 136-148. Available at: https://doi.org/10.1111/ijsw.12176
COUNTRY	Germany
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Return to unemployment
NON-SURVIVORS INCLUDED?	No.
DATA SOURCES	Administrative data: Data were obtained from the Department of Statistics of the German Federal Employment Agency. 1 206 recepients of start-up subsidy in 2005 (treated) and a control group of 224 641 non-supported individuals (other unemployed individuals).
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) The authors estimate individual-level effects from t+2 to t+6
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of area and selection bias. The authors provide estimates across German regions and across various individual characteristics.
KEY FINDINGS	The authors find positive and long-term effects on the probability of being employed or self-employed, i. e. economically active (in other words non-returns to unemployment).
PROGRAMME EXPENDITURE	169.66 mil. EUR allocated through the programme during 2007-2012 in Germany.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The results of the paper were included in the annual reports of the Institute for Employment Research concerning research on the system of “Basic Income Support for Job-Seekers” (Grundsicherung für Arbeitsuchende). The authors also wrote to inform the German Federal Ministry of Labour and Social Affairs. However, the authors cannot confirm that the government incorporated any of their recommendations.

Table B.40. New evidence on long-term effects of start-up subsidies: Matching estimates and their robustness
TABLE REFERENCE	B40
PROGRAMME NAME	German start-up subsidy (SUS) programme Gründungszuschuss
DATES	Years when the programme was operating: 2012-ongoing Evaluation period: 2012-2015 Year of the report: 2019 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to activate unemployed individuals through start-up subsidy and encourage them to become self-employed.
	Yes	Reduction of unemployment, increase in start-ups
EVALUATION THEME	Inclusive entrepreneurship; Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	Unemployed individuals
SOURCE OF EVIDENCE	Academic article: (Caliendo and Tübbicke, 2020[41]). New evidence on long-term effects of start-up subsidies: matching estimates and their robustness. Empirical Economics, (forthcoming). Available at: https://doi.org/10.1007/s00181-019-01701-9
COUNTRY	Germany
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Self- or regular employment, earnings
NON-SURVIVORS INCLUDED?	No
DATA SOURCES	Administrative and survey data: Data were obtained from the Integrated Labor Market Biographies (IEB) of the Federal Employment Agency (FEA) and from the follow-up surveys. 1 248 recepients of start-up subsidy in 2012 (20 000 treated indviduals in 2012 in total) and a control group of 1 204 non-supported individuals (other unemployed individuals).
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM), instrumental variables (IV) approach The authors estimate individual-level effects in t+20 months and t+40 months
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of area and selection bias.
KEY FINDINGS	The authors find positive and long-term effects on income and the probability of being employed or self-employed, i. e. economically active.
PROGRAMME EXPENDITURE	268 mil. EUR
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The authors have not presented their findings to the policymakers.

Table B.41. The outcome of coaching and training for self-employment. A statistical evaluation of outside assistance support programmes for unemployed business founders in Germany
TABLE REFERENCE	B41
PROGRAMME NAME	German start-up subsidy programme Überbrückungsgeld (Bridging Allowance)
DATES	Years when the programme was operating: 1986-ongoing Evaluation period: 2000-2005 Year of the report: 2015 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to activate unemployed individuals through a start-up subsidy (that could have been combined with coaching and training activities, and with discretionary start-up support) and encourage them to become self-employed.
	Yes	Reduction of unemployment, increase in start-ups
EVALUATION THEME	Inclusive entrepreneurship, business advice, coaching, mentoring and counselling, finance
INTERVENTION TYPE	Hard & Soft
TARGET GROUPS	Unemployed individuals
SOURCE OF EVIDENCE	Academic article: (Oberschachtsiek, 2015[42]). The outcome of coaching and training for self-employment. A statistical evaluation of outside assistance support programs for unemployed business founders in Germany. Journal for Labour Market Research, 48(1), 1-25. Available at: https://doi.org/10.1007/s12651-014-0161-6
COUNTRY	Germany
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Survival
NON-SURVIVORS INCLUDED?	Yes, firm survival is the main outcome variable.
DATA SOURCES	Administrative data: Data were obtained from the Integrated Employment Biographies (IEB) and from the Institute for Employment Research of the German Federal Employment Agency. The authors work with the four treated groups of self-employed individuals supported during 2000-2003, and they divide them according to the intensity of support received. The control group is composed of individuals receiving only the start-up subsidy (bridging allowance). The numbers of treated individuals are reported as follows: 209 040 individuals supported by start-up subsidy (of 418 856 individuals supported in total); 1 983 individuals who received an additional training (of 2 131 individuals supported in total), 10 107 individuals who received an extra coaching (of 13 737 individuals supported in total), and 17 790 individuals who received an extra discretionary start-up support (of 30 481 individuals supported in total).
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) The authors estimate firm-level effects in t+3
EVALUATION QUALITY SCORE	5
RELIABILITY COMMENTS	Analysis tackles the issue of area and selection bias. The authors provide estimates across German regions and across gender. Large sample study. However, no group of non-applicant entrepreneurs (or other unemployed individuals) is used as an additional control group.
KEY FINDINGS	The authors do not find conclusive evidence that the extra forms of support (in addition to the start-up subsidy) significantly increase the survival rates of subsidised businesses. The authors find empirical support for this assumption only for some forms of extra support and only in selected regions.
PROGRAMME EXPENDITURE	N/A
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	N/A

Table B.42. The ambiguous effects of public assistance to youth and female start-ups between job creation and entrepreneurship enhancement
TABLE REFERENCE	B42
PROGRAMME NAME	Start-up Programme Fare impresa (Doing Business)
DATES	Years when the programme was operating: 2011-2015 Evaluation period: 2011-2015 Year of the report: 2019 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to support youth and female business owners and unemployed individuals starting up in entrepreneurship in their early stages through allocation of soft loans and credit guarantees.
	Yes	Improved access to finance, reduction of unemployment
EVALUATION THEME	Inclusive entrepreneurship; Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	Youth, females, and unemployed individuals, irrespective of age or gender, with reference to a wide range of economic activities in the manufacturing, trade and tourism sectors
SOURCE OF EVIDENCE	Academic article: (Mariani et al., 2019[43]). The ambiguous effects of public assistance to youth and female start-ups between job creation and entrepreneurship enhancement. Scienze Regionali, 18(2), 237-260. Available at: https://www.rivisteweb.it/doi/10.14650/93649
COUNTRY	Italy
REGIONAL/LOCAL	One region study, programme focused on new businesses in Tuscany region
PERFORMANCE METRICS	Employment, firm survival
NON-SURVIVORS INCLUDED?	Yes, firm survival is one of the outcome variables
DATA SOURCES	Administrative data: Programme data and data from the regional job information system obtained from the regional government combined with the data obtained from the Business Register maintained by the Chambers of Commerce. 1 837 firms have received a credit guarantee in total. Out of these, 1 563 firms received, in addition, a subsidised soft loan. Firms that received also a soft loan were considered to be treated (1 563 firms), and the remaining were used as a control group (274 firms).
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) in combination with survival analysis The authors estimate firm-level effects from t+1 to t+3
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of industry bias and area bias. Separate estimates for youth and female start-ups are reported. Different kinds of treatment groups are, however, mixed together. An additional control group obtained from the population of non-applying firms is missing.
KEY FINDINGS	The authors find positive effects on firm survival and, to some extent, on further employment creation. Nevertheless, the positive effect on survival vanishes before the guaranteed loan is fully reimbursed.
PROGRAMME EXPENDITURE	N/A
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	Results were presented to the regional government and also to the wider public in a public event organised by the government. Regional start-up programmes launched after this evaluation offer microcredit in combination with a voucher to buy specialised services from consultants, which could raise managerial abilities and thus improve the quality/prospects of the supported projects.

Table B.43. Evaluation of the Spanish flat rate for young self-employed workers
TABLE REFERENCE	B43
PROGRAMME NAME	Social security reduction programme for youth self-employment from unemployment
DATES	Years when the programme was operating: 2013-ongoing Evaluation period: 2013-2014 Year of the report: 2017 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to activate unemployed individuals and specifically youth through social security contributions deduction and encourage them to become self-employed.
	Yes	Reduction of unemployment, increase in start-ups
EVALUATION THEME	Inclusive entrepreneurship; Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	Unemployed youth individuals
SOURCE OF EVIDENCE	Academic article: (Cueto, Mayor and Suárez, 2017[44]). Evaluation of the Spanish flat rate for young self-employed workers. Small Business Economics, 49(4), 937-951. Available at: https://doi.org/10.1007/s11187-017-9853-y
COUNTRY	Spain
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Firm survival
NON-SURVIVORS INCLUDED?	Yes, firm survival is the main outcome variable.
DATA SOURCES	Administrative data: Data were obtained from the Continuous Sample of Working Lives (CSWL, Muestra Continua de Vidas Laborales), an administrative dataset maintained by the Spanish Ministry of Employment and Social Security. 2 927 newly established youth self-employed in 2013 (about 50 000 treated individuals in 2013 in total) and a control group of 6 664 non-supported individuals (other newly established self-employed individuals in 2013).
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) in combination with a difference-in-differences (DiD) approach The authors estimate firm-level survival effects in t+1
EVALUATION QUALITY SCORE	5
RELIABILITY COMMENTS	Analysis tackles the issue of area and selection bias.
KEY FINDINGS	The authors find no statistically significant effect on firm survival.
PROGRAMME EXPENDITURE	N/A
MACRO IMPACT	The programme has significantly contributed to the increase of youth self-employment, as documented by the increase in rates of newly-established self-employed individuals. Nevertheless, this increase was followed by an increase in business closure rates.
POLICY IMPACT OF THE EVALUATION	The authors have presented results to the government and published a newspaper article about the results of the evaluation. However, they are not aware of any specific impact of the evaluation.

Table B.44. Is starting a business a sustainable way out of unemployment? Treatment effects of the Swedish start-up subsidy
TABLE REFERENCE	B44
PROGRAMME NAME	The Swedish Start-up Grants programme (SEP Programme)
DATES	Years when the programme was operating: 1984 (nowadays) Evaluation period: 2003-2007 Year of the report: 2016 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to activate unemployed individuals through public support (in the form of a non-repayable grant) and encourage them to become self-employed.
	Yes	Reduction of unemployment, increase in start-ups
EVALUATION THEME	Inclusive entrepreneurship; Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	Unemployed individuals
SOURCE OF EVIDENCE	Academic article: (Behrenz, Delander and Månsson, 2016[45]). Is starting a business a sustainable way out of unemployment? Treatment effects of the Swedish start-up subsidy. Journal of Labor Research, 37(4), 389-411. Available at:https://doi.org/10.1007/s12122-016-9233-4
COUNTRY	Sweden
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Probability of leaving unemployment
NON-SURVIVORS INCLUDED?	No
DATA SOURCES	Administrative data: Data were obtained from the Swedish Social Insurance Agency. 15 106 entrepreneurs supportedin 2003 in Sweden and a control group of 466 691 unemployed individuals who were not supported by any instrument of active labour market policy (non-applicants).
STEP LEVEL	6
METHODS	Panel data approach Propensity score matching (PSM) with a difference-in-differences approach The authors estimate effects in t+3 and in t+5
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of area and selection bias. The authors provide estimates across different levels of education attainment.
KEY FINDINGS	The authors find that participation in the programme has increased the probability of leaving unemployment. The results by educational attainment levels showed the largest effects for low educated unemployed.
PROGRAMME EXPENDITURE	The total programme expenditures were 800 mil. SEK annually from 2003 on and 500 mil. SEK annually from 2007 on (1 EUR = 10.7 SEK).
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The results were presented to representatives of the government (Swedish NAO) and based on the authors’ recommendations, the programme received more funding and expanded from 2016.

Regional and Local Focus

Table B.45. Publicly funded prestart support for new firms: who demands it and how it affects their employment growth
TABLE REFERENCE	B45
PROGRAMME NAME	Pre-start support via a Funded Business Development Centre (PFBDC)
DATES	Years when the programme was operating: 2002-2005 Evaluation period: 2000-2005 Year of the report: 2011 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to support new business formation in Navarra region. Individuals interested in starting a business could ask for public support for business start-up through a Funded Business Development Centre (PFBDC). Depending on the stage of the business development, they could obtain counselling, training, a place in an incubator or a direct financial subsidy.
	Yes	Increase in start-ups
EVALUATION THEME	Regional and local; Business advice, coaching, mentoring and counselling; Finance
INTERVENTION TYPE	Hard & Soft
TARGET GROUPS	New firms in Navarra region
SOURCE OF EVIDENCE	Academic article: (Capelleras, Contín-Pilart and Larraza-Kintana, 2011[46]). Publicly funded prestart support for new firms: who demands it and how it affects their employment growth. Environment and Planning C: Government and Policy, 29(5), 821-847. Available at:https://doi.org/10.1068/c10110b.
COUNTRY	Spain
REGIONAL/LOCAL	One region study, programme focused on new businesses in Navarra region
PERFORMANCE METRICS	Employment
NON-SURVIVORS INCLUDED?	No
DATA SOURCES	Survey and administrative data: Survey data collected in 2001 and 2005 in combination with the data collected from the Government of Navarra. 78 firms supported (100% of population), 114 non-supported firms (non-applicants).
STEP LEVEL	6
METHODS	Panel data approach Two-stage Heckman selection model The authors estimate firm-level effects in t+5
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of area and selection bias. Three different kinds of support are analysed and the separate findings are reported. The authors do not report the total number of supported start-ups by the government.
KEY FINDINGS	The authors find positive effects on employment growth only in the case of soft business support (i.e. knowledge development support), and not in the case of tangible public support.
PROGRAMME EXPENDITURE	N/A
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The authors have presented the results to the government and to the relevant stakeholders. The outcomes were considered during the establishment of the First Plan for Entrepreneurship in Navarra (I Plan de Emprendimiento de Navarra), which also included entrepreneurship training activities as pointed out by the research study.

Support in Areas of Disadvantage

Table B.46. Public investment subsidies and firm performance – Evidence from Germany
TABLE REFERENCE	B46
PROGRAMME NAME	Improving regional economic structures (Verbesserung der regionalen Wirtschaftsstruktur - GRW)
DATES	Years when the programme was operating: 2007-2013 Evaluation period: 2007-2014 Year of the report: 2018 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 Programme aimed to mitigate regional disparities through investment subsidies allocated to firms. All subsidised investment projects had to generate employment in the region where the project was realised.
	Yes	Increase in employment, wealth and competitiveness
EVALUATION THEME	Support in areas of disadvantage; Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	Firms in East Germany (lagging regions)
SOURCE OF EVIDENCE	Academic article: (Brachert, Dettmann and Titze, 2018[47]). Public investment subsidies and firm performanceEvidence from Germany. Jahrbcher fr Nationalkonomie und Statistik, 238(2), 103-124. Available at: https://doi.org/10.1515/jbnst-2017-0131
COUNTRY	Germany
REGIONAL/LOCAL	One region study, the evaluation focused on manufacturing firms in Saxony-Anhalt region
PERFORMANCE METRICS	Employment, turnover, gross fixed capital, labour productivity
NON-SURVIVORS INCLUDED?	No
DATA SOURCES	Statistical office data and administrative data: Annual surveys provided by the German Research Data Centre within the AFiD database in combination with GRW programme data. 254 firms supported in 2007 in Saxony-Anhalt region (of 1 208 supported in total during 2007-2013), 19 821 non-supported firms from West German regions (no access to funding).
STEP LEVEL	6
METHODS	Panel data approach Coarsened exact matching (CEM) in combination with a fixed-effects difference-in-differences (FEDiD) approach The authors estimate firm-level effects from t+1 to t+6
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of area and industry bias.
KEY FINDINGS	The authors find positive short- and medium-run effects on firm employment. The effects on firm turnover remain significant and positive only in the medium-run. Gross fixed capital formation responds positively to GRW funding only during the mean implementation period of the projects but becomes insignificant afterwards. Finally, the effect of GRW funding on labour productivity remains insignificant throughout the whole period of analysis.
PROGRAMME EXPENDITURE	7.422 billion Euro allocated through GRW during 2007-2013 in Germany,1.377 billion Euro allocated to Saxony-Anhalt region.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The results were presented to the government and disseminated to relevant stakeholders in Germany, but also in the European Commission (Directorate-General for Regional and Urban Policy). The authors say that based on their findings, the Federal State of Thuringia has changed the guidelines on the GRW-funding. The policymakers incorporated the recommendation to make a change in policy goal, i.e. to shift from to the goal of job creation to productivity issues in the funding.

Table B.47. How are growth and productivity in private firms affected by public subsidy? Evidence from a regional policy
TABLE REFERENCE	B47
PROGRAMME NAME	Regional policy determined by the Law 488/1992 (L.488)
DATES	Years when the programme was operating: 1996-2007 Evaluation period: 1996-2004 Year of the report: 2011 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to boost private investments in industrial structure development and job creation in less developed regions through the allocation of investment subsidies.
	Yes	Higher competitiveness, increase in employment
EVALUATION THEME	Support in areas of disadvantage; Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	Manufacturing and extractive firms doing business in lagging regions
SOURCE OF EVIDENCE	Academic article: (Bernini and Pellegrini, 2011[48]). How are growth and productivity in private firms affected by public subsidy? Evidence from a regional policy. Regional Science and Urban Economics, 41(3), 253-265. Available at: https://doi.org/10.1016/j.regsciurbeco.2011.01.005
COUNTRY	Italy
REGIONAL/LOCAL	One country study, the programme focused on manufacturing and extractive firms doing business in lagging regions
PERFORMANCE METRICS	Employment, sales, fixed assets, value added per labour costs, debt costs, total factor productivity (TFP)
NON-SURVIVORS INCLUDED?	No
DATA SOURCES	Administrative and commercial data: Firm-level data from the AIDA database in combination with the programme data. 574 firms supported by subsidies during 1996-1998 (665 firms were supported in total) and 848 non-supported firms (rejected-applicants).
STEP LEVEL	5
METHODS	Panel data approach Propensity score matching (PSM) in combination with a difference-in-differences (DiD) approach The authors estimate firm-level effects in t+1
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of area and industry bias. The results are not compared with the additional control group of non-applicants and the analysis lacks data for small firms participating in the programme.
KEY FINDINGS	The authors find positive effects on sales, value added, employment and fixed assets, but negative effects on total factor productivity.
PROGRAMME EXPENDITURE	The total amount of public resources allocated through subsidies to the firms during 1996-2007 was 23 bil. EUR.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The authors did not present the results to the government.

Table B.48. Do subsidies to private capital boost firms' growth? A multiple regression discontinuity design approach
TABLE REFERENCE	B48
PROGRAMME NAME	Regional policy determined by the Law 488/1992 (L.488)
DATES	Years when the programme was operating: 1996-2007 Evaluation period: 1995-2004 Year of the report: 2014 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to boost private investments in industrial structure development and job creation in less developed regions through the allocation of investment subsidies.
	Yes	Higher competitiveness, increase in employment
EVALUATION THEME	Support for disadvantaged areas; Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	Manufacturing and extractive firms doing business in lagging regions
SOURCE OF EVIDENCE	Academic article: (Cerqua and Pellegrini, 2014[49]). Do subsidies to private capital boost firms' growth? A multiple regression discontinuity design approach. Journal of Public Economics, 109, 114-126. Available at: https://doi.org/10.1016/j.jpubeco.2013.11.005
COUNTRY	Italy
REGIONAL/LOCAL	One country study, the programme focused on manufacturing and extractive firms doing business in lagging regions. Evaluation focused on southern regions.
PERFORMANCE METRICS	Employment, sales, fixed assets, value added per labour costs
NON-SURVIVORS INCLUDED?	No
DATA SOURCES	Administrative and commercial data: Firm-level data from the AIDA database in combination with programme data from the Ministry of Economic Development. 428 firms supported by subsidies during 1996-1998 in south Italy (1 784 firms were supported in total) and 531 non-supported firms (rejected applicants).
STEP LEVEL	6
METHODS	Panel data approach Regression discontinuity design (RDD) The authors estimate firm-level effects from t+1 to t+6
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of area and industry bias.
KEY FINDINGS	The evaluation finds positive effects on tangible assets, turnover and employment, but insignificant effects on value added per labour costs (labour productivity).
PROGRAMME EXPENDITURE	The total amount of public resources allocated through subsidies to the firms during 1996-2007 was 23 bil. EUR.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	N/A

Table B.49. Industrial policy evaluation in the presence of spillovers
TABLE REFERENCE	B49
PROGRAMME NAME	Regional policy determined by the Law 488/1992 (L.488)
DATES	Years when the programme was operating: 1996-2007 Evaluation period: 1995-2001 Year of the report: 2017 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 The programme aimed to boost private investments in industrial structure development and job creation in less developed regions through the allocation of investment subsidies.
	Yes	Higher competitiveness, increase in employment
EVALUATION THEME	Support in areas of disadvantage; Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	Manufacturing and extractive firms doing business in lagging regions
SOURCE OF EVIDENCE	Academic article: (Cerqua and Pellegrini, 2017[50]). Industrial policy evaluation in the presence of spillovers. Small Business Economics, 49(3), 671-686. Available at: https://doi.org/10.1007/s11187-017-9855-9
COUNTRY	Italy
REGIONAL/LOCAL	One country study
PERFORMANCE METRICS	Employment, sales, fixed assets, total factor productivity (TFP)
NON-SURVIVORS INCLUDED?	No
DATA SOURCES	Administrative and commercial data: Firm-level data from database AIDA, in combination with the programme data. 213 firms supported by subsidies during 1996-1998 (665 firms were supported in total), 693 rejected applicant firms and 1 352 non-supported firms (non-applicants).
STEP LEVEL	6
METHODS	Panel data approach Coarsened exact matching (CSM) in combination with a difference-in-differences (DiD) approach The authors estimate firm-level effects from t+1 to t+3
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of area and industry bias. The authors also report results for firms located around the treated firms to observe possible spillover effects.
KEY FINDINGS	The authors find positive effects on fixed assets, sales and employment and negative effects on TFP.
PROGRAMME EXPENDITURE	The total amount of public resources allocated through subsidies to the firms during 1996-2007 was 23 bil. EUR.
MACRO IMPACT	N/A
POLICY IMPACT OF THE EVALUATION	The authors have not presented results to the government, because the policy under analysis was phased out from 2007 onwards.

Table B.50. Some causal effects of an industrial policy
TABLE REFERENCE	B50
PROGRAMME NAME	Regional Selective Assistance (RSA) Programme
DATES	Years when the programme was operating: 1972-ongoing Evaluation period: 1997-2004 Year of the report: 2019 (Published)
STATED OBJECTIVES Programme goal stated Final objectives available	Yes	Objective specification score: 2 Programme aimed to increase employment through investment subsidies (i.e. grants for industrial development) allocated to firms doing business in manufacturing located in lagging regions.
	Yes	Increase in employment, increase in competitiveness
EVALUATION THEME	Support in areas of disadvantage; Finance
INTERVENTION TYPE	Hard
TARGET GROUPS	Manufacturing firms located in lagging regions
SOURCE OF EVIDENCE	Academic article: (Criscuolo et al., 2019[51]). Some causal effects of an industrial policy. American Economic Review, 109(1), 48-85. Available at: https://doi.org/10.1257/aer.20160034
COUNTRY	United Kingdom
REGIONAL/LOCAL	One country study, the programme focused on manufacturing firms in lagging regions
PERFORMANCE METRICS	Employment, investments, value-added per employee, total factor productivity (TFP)
NON-SURVIVORS INCLUDED?	No
DATA SOURCES	Administrative data: Data were provided by the UK Census Bureau (Office of National Statistics), Selective Assistance Management Information System (SAMIS) database, the Interdepartmental Business Register (IDBR), and the Annual Respondents Database (ARD). 4 550 firms supported during 1997-2000 (82% of all RSA recipients), 39 308 non-supported firms (non-applicants).
STEP LEVEL	6
METHODS	Panel data approach Instrumental variables (IV) approach The authors estimate firm-level effects in t+3
EVALUATION QUALITY SCORE	4
RELIABILITY COMMENTS	Analysis tackles the issue of area and industry bias. Effects across firm size are reported.
KEY FINDINGS	The authors find positive effects of the programme on employment and investments, but no effects on total factor productivity (TFP). They also found heterogeneous effects across firm sizes. Statistically significant effects are reported for small firms, but smaller or no effects for large firms.
PROGRAMME EXPENDITURE	164 mil. GBP allocated through RSA during 1997-2004.
MACRO IMPACT	The authors report the overall increased employment in lagging regions and reduction of unemployment.
POLICY IMPACT OF THE EVALUATION	The authors have presented results to UK and European Union (EU) policymakers. The findings of this study have influenced the governance of state aid investments at the EU for large firms. The authors refer to the following two documents: (European Commission, 2014[52]). Common methodology for state aid evaluation. Brussels. Available at: http://ec.europa.eu/competition/state_aid/modernisation/state_aid_evaluation_methodology_en.pdf (European Commission, 2019[53]). Explanatory note on the paper of the services of DG Competition containing draft regional aid guidelines 2014-2020. [online] Available at: http://ec.europa.eu/competition/consultations/2013_regional_aid_guidelines/explanatory_note_en.pdf