WP 3 - Innovation and Access to Finance Project Steering Meeting and Stakeholders Meeting 29-30 September 2016 Venue: Ekonomski institut, Zagreb (EIZ)/The Institute of Economics, Zagreb Michele CINCERA, ULB-Solvay-iCite (Belgium) Anabela SANTOS, ULB-Solvay-iCite (Belgium) This project is co-funded by the European Union
Outline Focus of WP3 Commitment 10. Put in place EU level financial instruments to attract private finance Commitment 11. Ensure cross-border operation of venture capital funds Commitment 12. Strengthen cross-border matching of innovative firms with investors Commitment 13. Review State Aid Framework for RDI Integration in NEMESIS model What we plan to do? Presentation of first results 2
Integration in NEMESIS model Commitment # Category Thematic Inputs Commitment 10 Selected Finance Data and elasticities Commitment 11 Selected Finance Data and elasticities Commitment 12 Selected Finance Data and elasticities Commitment 13 Selected Finance Data and elasticities Why? Precise objectives Significant direct impact on the European innovation system. Data are available (under some restrictions) and it is possible to assess its direct impact. 3
Commitment 10. Put in place EU level financial instruments to attract private finance Commitment rationale Access to finance for innovative firms R&D investment productivity Innovation Proposed solution Putting in place FINANCIAL INSTRUMENTS to support investment in the early stages of start-up development To enhance VENTURE CAPITAL investment for fast growing firms To ensure ACCESS TO LOANS for innovative fast growing SMEs Economic growth 4
Commitment 10. Put in place EU level financial instruments to attract private finance HOW TO INTEGRATE THE COMMITMENT IN NEMESIS? Model # Research Question Data Methodology Model 1.1. Model 1.2. Model 2.1. How macroeconomic conditions could influence RSFF allocation? Which are the effect of RSFF on country R&D expenditures and jobs? For which kind of firm is access to finance a pressing problem? Countries: EU28 Period: 2007 2013 Main source: EIB Countries: EU28 Period: 2007 2013 Main source: EIB Countries: EU12 Period: 2014 Source: SAFE micro dataset (ECB) Panel data regression model Y= Investment funded by RSFF (% GDP) X=macroeconomic indicators Stochastic Frontier Analysis R&D production function (input = RSFF) Macroeconomic indicators Ordered probit or logit Y= Degree of finance pressing problem X= firm characteristics and market conditions 5
Commitment 10. Put in place EU level financial instruments to attract private finance HOW TO INTEGRATE THE COMMITMENT IN NEMESIS? Model # Research Question Data Methodology Model 2.2. Model 2.3. Model 2.4. Is application (external source of finance) success the same for all kind of firm and for all sources of financing? Which factors influence the interest rate of a credit line? Could access to finance influence firm performance? Countries: EU12 Period: 2014 Source: SAFE micro dataset (ECB) Ordered probit or logit Y= Scale of application success X= firm characteristics and market conditions Fractional regression model Y= interest rate charged for the credit line or bank overdraft (= FIRM RISK) X= firm characteristics and market conditions Multinomial logit model or a binary model Y= change in turnover or profit (increased, remained unchanged or decreased) X= firm characteristics and market conditions 6
Commitment 10. Put in place EU level financial instruments to attract private finance HOW TO INTEGRATE THE COMMITMENT IN NEMESIS? Model # Research Question Data Methodology Model 2.5. How access to finance could influence R&D decision? Countries: EU12 Period: 2014 Source: SAFE micro dataset (ECB) Binary model Y= doing or not R&D X= firm characteristics and market conditions Comments: Innovative firms = purpose of external financing developing and launching a new product or services. Possibility to use SAFE micro dataset in panel (panel key received last week). 7
Market conditions (= degree of pressing problem) Firm characteristics Relevance of financial instruments Commitment 10. Put in place EU level financial instruments to attract private finance Model 2.5. R&D Decision and Access to Finance Y = Doing or not R&D based on the purpose of external financial > developing and launching a new product or services Variables Coef. Std. Err. dy/dx SME -0.969 *** 0.134-0.1250 Family ownership 0.096 0.059 0.0124 Firms ownership 0.115 0.089 0.0148 VC or BA ownership 0.516 * 0.296 0.0666 Old firm ( 10 years) -0.347 *** 0.068-0.0448 Export quota (%) 0.006 *** 0.001 0.0008 Customer (0-10) 0.051 *** 0.012 0.0066 Competition (0-10) -0.031 ** 0.013-0.0039 Finance (0-10) 0.016 0.010 0.0021 Cost (0-10) -0.022 0.013-0.0028 Staff (0-10) 0.041 *** 0.012 0.0053 Regulation (0-10) 0.004 0.011 0.0005 Variables Coef. Std. Err. dy/dx Internal funds 0.241 *** 0.060 0.0311 Grants 0.330 *** 0.061 0.0426 Credit line -0.016 0.062-0.0021 Bank loan -0.074 0.063-0.0096 Trade credit -0.003 0.062-0.0003 Equity 0.209 *** 0.076 0.0270 Debt securities 0.156 0.114 0.0201 Leasing 0.113 ** 0.058 0.0146 Factoring 0.147 0.080 0.0190 Constant -2.308 *** 0.214 Observations 10,914 Log likelihood -4,545.01 Pseudo R2 0.0775 Correctly classified 83,32% Source: Authors own elaboration. Note: Results of LOGIT regression model *** p<0.01, ** p<0.05, * p<0.1 The regression includes 4 activity sector dummies and 12 country dummies. 8
Commitment 11. Ensure cross-border operation of venture capital funds Commitment rationale Easy cross-border investment Proposed solution Removing obstacles and improving the fiscal environment of VC R&D investment Innovation Economic growth productivity 9
Commitment 11. Ensure cross-border operation of venture capital funds HOW TO INTEGRATE THE COMMITMENT IN NEMESIS? Model # Research Question Data Methodology Model 3.1. Model 3.2. How macroeconomic conditions can affect the cross-border operation of venture capital funds? How could cross-border venture capital funds impact on innovation, competitiveness and growth? Countries: EU20 Period: 2007 2015 Main source: EVCA and EUROSTAT Countries: EU20 Period: 2007 2015 Main source: EVCA and EUROSTAT Panel data regression model Y= VC investment by foreign firms X=macroeconomic indicators Stochastic Frontier Analysis Production function of R&D or VA or GDP (input = foreign VC investment) Macroeconomic indicators 10
Commitment 11. Ensure cross-border operation of venture capital funds Model 3.1. Cross-border VC and macroeconomic conditions Y = Share of foreign VC investment (% of total VC investment in country i) Methodology: Panel data with fractional regression model (GLM and logit) Foreign Direct Investment (Walsh and Yu, 2010) Market size and growth potential Degree of openness of the host economy Exchange Rate Valuation Political stability Quality of institutions and level of development State of cluster development Foreign Venture Capital Investment Pandya and Leblang (2011) GDP Distance Common Official Language Shared Common Border Common Colonial History Common Legal Origin Common Currency Patents Stock Market Development Capital Account Openness Dual Taxation Treaty Share of Migrants with Graduate Education Index of Political Constraints Democracy Score R&D investments Becker (2013) Access to finance Degree of competition Public support to R&D Concentration of skilled workers Spillovers from foreign R&D 11
Commitment 11. Ensure cross-border operation of venture capital funds Model 3.1. Cross-border VC and macroeconomic conditions Y = Share of foreign VC investment (% of total VC investment in country i) Methodology: Panel data with fractional regression model (GLM and logit) Variables Coef. Std. Err. dy/dx L.inv_protect 4.393 * (2.470) 0.81 L.inv_protect_2-0.439 * (0.226) -0.08 erp 187.5 *** (52.04) 34.76 erp2-1,756 *** (536.6) -325.45 erp3 4,670 *** (1,553) 865.67 L.foreign_own 0.86 ** (0.396) 0.16 L.eberd_abroad -228 ** (90.01) -42.25 D.openess_index 2.488 ** (1.289) 0.46 L.var_product 1.596 (3.847) 0.30 estock_trad1-0.597 * (0.310) -0.11 L.pcr_pv -1.438 (1.914) -0.27 D.subs_gdp 95.59 *** (36.41) 17.72 Observations 180 Log pseudolikelihood -71.342 Source: Authors own elaboration. Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 The regression includes sector and time fixed effects Positive impact Strength of investor protection index (0-10) Prevalence of foreign ownership in the country (0-7) Total Equity Risk Premium (%) Openness index (export and import % GDP) Subsidy (% GDP) Negative impact BERD funded by abroad (% GDP) Stock traded (% GDP) Not significant Labor productivity Political stability 12
Commitment 12. Strengthen cross-border matching of innovative firms with investors Commitment rationale Proposed solution More efficient economic ecosystem Easier matching investors and innovative firms Improving access to finance for innovative-firms productivity Economic growth Matching the supply and demand sides for innovative projects and ideas Developing a new financial culture among entrepreneurs Promoting the cooperation among Business Angels and Venture Capitalists Entry in NEMESIS Difficult to quantify it and there is no data available 13
Commitment 12. Strengthen cross-border matching of innovative firms with investors HOW TO INTEGRATE THE COMMITMENT IN NEMESIS? Model # Research Question Data Methodology Model 4.1. Model 4.2. How innovation capacity can be affected by the network and collaboration across entities? How economic growth and productivity can be affected by the network and collaboration across entities? Countries: EU28 Period: 2011 2015 Main source: Global Innovation Index (GII) and EUROSTAT Stochastic Frontier Analysis R&D Production function (input = Private R&D expenditure funded from abroad) Inefficiency factors: Innovation linkages indicators from GII Stochastic Frontier Analysis Production function (Labor productivity or GDP) Inefficiency factors: Innovation linkages indicators from GII Entities = e.g. firms/universities or firms/vc 14
Commitment 13. Review State Aid Framework for RDI Commitment rationale Proposed solution More effective legal framework for RDI State Aids Innovation and employment Economic growth and sustainability Simplification of RDI state aid Revision of legal framework for RDI State aids in 2008. State Aid Modernization (SAM) and the new General Block Exemption Regulation (GBER) launched in 2012. 15
Commitment 13. Review State Aid Framework for RDI HOW TO INTEGRATE THE COMMITMENT IN NEMESIS? Model # Research Question Data Methodology Model 5. How legal framework could affect the effectiveness and efficiency of RDI state aids? Countries: EU28 Period: 2006 2014 Main source: EUROSTAT and Global Competitiveness Index Stochastic Frontier Analysis R&D Production function Dummies variables for 2008 and 2012 (years when legal framework was revised/changed) Inefficiency variables: regulatory framework including bureaucracy 16
Thank you Michele CINCERA mcincera@ulb.ac.be Anabela SANTOS asantos@ulb.ac.be 17
STOCHASTIC FRONTIER ANALYSIS (SFA) The methodology requires the estimation of a production function. Inefficiency = difference between the optimal and observed levels of outputs. Figure 2. The stochastic production frontier model Source: Farrell (1957). 18
Commitment 10. Put in place EU level financial instruments to attract private finance Positive impact on doing R&D Negative impact on doing R&D VC or BA ownership (vs other type): increase the probability in 6.7 p.p. Customer and staff pressing problem: 1 unit increase the probability in 0.7 p.p. and 0.4 p.p, respectively. Relevance of different financing source (internal funds, grants and equity) increase the probability in 3.1p.p., 4.3 p.p. and 2.7 p.p., respectively. Firm size [SME vs non-sme] decrease the probability in 12.5 p.p. Firm age [old firm ( 10 years) vs less 10 years] decrease the probability in 4.5 p.p. Competition pressing problem: 1 unit decrease the probability in 0.4 p.p. 19
Empirical model Conceptual framework Commitment 13. Review State Aid Framework for RDI Macroeconomic factors - e.g. regulatory framework, national resources (human and infrastructure), collaboration, access to finance, economic climate, demand side, etc. OBJECTIVE INPUT Allocative Efficiency Technical Efficiency OUTPUT Effectiveness OUTCOME Increase private R&D investment Public R&D: Private R&D funding by the government R&D in Government sector R&D in Higher Education sector R&D tax credit (B-Index) Private R&D spending funding by Private Sector Competitiveness Productivity Source: Authors own elaboration based on Mandl et al. (2008:3), EC (2008:442) and Cincera et al. (2009:22). 20