Working Party No. 1 on Macroeconomic and Structural Policy Analysis

For Official Use ECO/CPE/WP1(2005)2/ANN1 ECO/CPE/WP1(2005)2/ANN1 For Official Use Organisation de Coopération et de Développement Economiques Organisation for Economic Co-operation and Development 24-Feb-2005 English - Or. English ECONOMICS DEPARTMENT ECONOMIC POLICY COMMITTEE Working Party No. 1 on Macroeconomic and Structural Policy Analysis INNOVATION POLICIES: INNOVATION IN THE BUSINESS SECTOR ANNEX 1: TABLES AND FIGURES This document has been prepared by the Economics Department for Working Party No. 1 of the Economic Policy Committee. Contacts: Nigel Pain, tel: (33-1) 45 24 91 54; fax: (33-1) 44 30 63 78; email: Nigel.Pain@oecd.org; Florence Jaumotte, tel: (33-1) 45 24 14 06; fax: (33-1) 44 30 63 83; email: Florence.Jaumotte@oecd.org. English - Or. English JT00179240 Document complet disponible sur OLIS dans son format d'origine Complete document available on OLIS in its original format

ANNEX 1 TABLES AND FIGURES Tables 1.1 Long-run effects of a one standard deviation increase in policy and framework factors 2.1 Comparison of innovation performance by various criteria, 2001 2.2 Comparison of macroeconomic indicators and survey-based indicators of innovation performance, 1998-2000 Figures 1.1 Contribution of science policies and framework conditions to cross-country differences in business R&D intensity, 2000 1.2 Contribution of science policies and framework conditions to growth in R&D intensity, 1991-2000 2.1 Business sector R&D intensity 2.2 Convergence in business sector R&D intensity 2.3 Business and non-business R&D intensity, 1996-2000 2.4 Non-business R&D intensity 2.5. Business sector researchers 2.6 Composition of innovation spending 2.7 Total patents per million of working age population 2.8 Triadic patents per million of working age population 2.9 Triadic patents per million of business R&D spending 2.10 Choice of protection method 2.11 Comparison of countries innovative performance by various measures, 1998-2000 2.12 Share of new products in turnover 2.13 Pure innovation and imitation 3.1 The state of tax and subsidisation policies 3.2 Changes in tax and subsidisation policies 3.3 Business funding of non-business sector R&D 3.4 Convergence in intellectual property rights (Park index) 4.1 Financial factors 4.2 Evolution of the share of equity financing in financial development 4.3 PMR index 4.4 Index of FDI restrictions 2

Table 1.1 Long-run effects of a one standard deviation increase in policy and framework factors 1 Measured in percentage change of the dependent variable Business R&D spending Total domestic patents Science policies and institutions B-index 2-1¾ -6 Subsidies for private R&D / GDP ratio ¼ -3 Share of business funding in non-business R&D 8¼ 2½ Non-business R&D / GDP ratio 7¼ 3¾ IPR index 1½ 8 real wage of researchers -3¼ -¾ Years of education 1 ¾ Economic conditions Profit / GDP ratio 5¼ 4¼ Private sector credit / GDP ratio -1½ -3¼ Equity financing / GDP ratio 5¾ 10 Foreign R&D stock / GDP ratio 12¾ 6 Openness -5¾ -4¼ Import penetration -¼ 0 Real interest rate -5-2¾ Real exchange rate -3-1¾ Framework policies (decrease) Product market regulation 9 4¼ FDI restrictions.. 13 Employment protection legislation 1 6½ 1. This table is taken from Table A3.8 of Annex 3. The standard deviation is the average of within-country standard deviations, and the effects of a one standard deviation increase in policy and framework factors are evaluated at the sample mean of the variables. 2. The B-index is defined as one minus the rate of tax subsidy for R&D. An increase in the B-index implies a less generous tax treatment of R&D. Source: OECD estimates 3

Table 2.1 Comparison of innovation performance by various criteria, 2001 1 Total R&D intensity Business R&D intensity Non-business R&D intensity Scientists share 2 R&D employees share 2 Triadic patents (1999) Average indicator 3 1 1 3 3 2 2 2.0 2 2 2 5 1 4 3.7 Japan 3 3 4 2 6 3 2.7 Iceland 4 6 1 4 5 13 7.0 United States 5 4 11 1 6 4.0 Switzerland 6 5 13 7 3 1 4.7 7 7 9 8 8 5 6.7 8 8 10 12 4 8 9.3 9 10 6 13 9 10 10.7 10 9 17 9 7 9 9.3 11 11 12 10 10 11 10.7 Canada 12 14 7 14 14 15 13.7 13 13 8 16 12 7 12.0 United Kingdom 14 12 14 15 15 12 13.7 15 15 15 6 11 14 11.7 Australia 16 17 5 17 16 16 16.3 Ireland 17 16 20 11 13 17 15.0 18 18 18 18 17 18 18.0 19 19 19 19 18 19 19.0 20 20 16 20 19 20 20.0 Correlation with Total R&D intensity 1 0.99 0.73 0.85 0.94 0.87 0.97 1. The comparison is based on rank orders according to the various criteria. Rankings are a rough measure of crosscountry differences. A more refined measure would use deviations from the country mean expressed in multiple of the standard deviation of countries' observations around the mean. Countries in the table are ordered by decreasing level of total R&D intensity. 2. The employment of scientists and R&D personnel is expressed as a share of total dependent employment. There are no data on the R&D employee share for the United States and the ranking according to this criterion is not perfectly comparable because only 19 countries are included instead of 20. 3. The average is the simple arithmetic average of the rankings for total R&D intensity, scientist share and triadic patents. Source: OECD Main Science and Technology Indicators database, R&D database and Patent database. 4

Table 2.2 Comparison of macroeconomic indicators and survey-based indicators of innovation performance, 1998-2000 1 Country Macroeconomic indicators CIS indicators of innovation spending and protection Business R&D intensity (2001) Triadic patents (1999) Proportion of firms engaging in intramural R&D Innovation spending index 2 Proportion of firms applying for a patent Aggregate protection index 3 CIS measures of implementation of innovation Proportion of successful innovators Share of new products in turnover 1 1 1 2 9 2 2 1 3 4 3 8 2 Iceland 3 10 9 10 14 14 2 12 4 3 2 1 3 5 1 1 5 5 4 5 8 10 7 7 6 6 3 2 6 6 3 6 7 7 6 7 2 7 10 10 8 8 5 4 5 8 United kingdom 9 9 11 1 14 10 4 5 6 9 9 6 9 11 11 7 8 7 8 12 11 12 12 10 9 10 11 11 4 13 13 11 11 12 13 13 3 14 14 8 4 13 12 4 5 Correlation with Business R&D intensity 1.00 0.81 0.69 0.42 0.57 0.46 0.38 0.00 Correlation with triadic patents 0.81 1.00 0.91 0.65 0.75 0.64 0.27 0.18 Correlation with % successful innovators 0.38 0.27 0.45 0.61 0.06-0.14 1.00 0.03 Correlation with share new products in turnover 0.00 0.18 0.27 0.44 0.16 0.25 0.03 1.00 Number of countries 14 14 11 11 14 14 14 12 1. Care has to be taken when interpreting cross-country comparisons made with the aggregated data in CIS, as there are differences in the sample size used in the respective national components of the survey. A number of indicators (in italics) are available only for a subset of countries so that the ranks of countries cannot be directly compared across all indicators, though their rank-ordering can be compared. Countries in the table are ordered by decreasing level of business R&D intensity. 2. The innovation spending index is calculated as the arithmetic average of the proportions of firms engaging in the various types of innovation spending. 3. The aggregate protection index is calculated as the arithmetic average of the proportions of firms using the various types of protection. Source: OECD Main Science and Technology Indicators database, R&D database and Patent database; Community Innovation Survey 3 (European Commission). 5

Figure 1.1 Contribution of science policies and framework conditions to cross-country differences in business R&D intensity, 2000 Measured in percentage point deviation of R&D intensity relative to average OECD R&D intensity 1 Business R&D intensity Science policies and institutions 2 Framework conditions 3 SWE FIN JPN CHE DEU DNK BEL FRA GBR AUT NLD CAN NOR IRE AUS ITA ESP PRT -1 0 1 2 SWE FIN JPN CHE DEU DNK BEL FRA GBR AUT NLD CAN NOR IRE AUS ITA ESP PRT -1 0 1 2 SWE FIN JPN CHE DEU DNK BEL FRA GBR AUT NLD CAN NOR IRE AUS ITA ESP PRT -1 0 1 2 1. The contributions have been calculated based on the long-run parameters from the equation reported in column 3 of Table A3.2. The OECD average refers to the unweighted geometric average of the OECD countries included in the sample (due to the logarithmic regression model). 2. Science policies include R&D tax incentives, subsidies for private R&D, business funding of non-business R&D, non-business R&D intensity, intellectual property rights, the share of scientists in total dependent employment and absorptive capacity (capacity to understand and make use of foreign knowledge). 3. Framework conditions include financial factors, real interest rates, real exchange rates, foreign exposure (foreign R&D stock and openness), import penetration, and product market regulation. Source: OECD estimates. 6

Figure 1.2 Contribution of science policies and framework conditions to growth in R&D intensity, 1991-2000 Measured in percentage point deviation of R&D intensity between 1991 and 2000 1 Business R&D intensity Science policies and institutions 2 Framework conditions 3 FIN SWE DNK AUT BEL CAN IRE AUS NLD PRT ESP JPN CHE NOR DEU ITA FRA GBR -0.5 0 0.5 1 1.5 FIN SWE DNK AUT BEL CAN IRE AUS NLD PRT ESP JPN CHE NOR DEU ITA FRA GBR -0.5 0 0.5 1 1.5 FIN SWE DNK AUT BEL CAN IRE AUS NLD PRT ESP JPN CHE NOR DEU ITA FRA GBR -0.5 0 0.5 1 1.5 1. The contributions have been calculated based on the long-run parameters from the equations reported in column 1 of Table A3-6 and in column 3 of Table A3.2. 2. Science policies include R&D tax incentives, subsidies for private R&D, business funding of non-business R&D, non-business R&D intensity, intellectual property rights and absorptive capacity (capacity to understand and make use of foreign knowledge). 3. Framework conditions include financial factors, real interest rates, real exchange rates, foreign exposure (foreign R&D stock and openness), import penetration, product market regulation, employment protection legislation, human capital and the domestic economy-wide average wage. Source: OECD estimates. 7

Figure 2.1 Business sector R&D intensity % of GDP, average per annum 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 Australia Ireland Iceland Canada UK Switzerland Japan 1996-2000 1981-85 2001 Source: OECD Main Science and Technology Indicators database and R&D database. 8

Figure 2.2 Convergence in business sector R&D intensity 2.0 1.5 1.0 0.5 FIN ISL DNK AUT AUS IRE CAN SWE JPN BEL Change 1981-2001 (% points) 0.0 NOR PRT ESP R 2 = 0.0833 FRA DEU ITA NLD CHE GBR -0.5 0.0 0.5 1.0 1.5 2.0 2.5 1981 level Source: OECD Main Science and Technology Indicators database and R&D database. 9

Figure 2.3 Business and non-business R&D intensity, 1996-2000 1 Average per annum 3.0 SWE R 2 = 0.3777 2.5 2.0 CHE JPN FIN BERD/GDP (%) 1.5 UK DEU BEL FRA DNK 1.0 IRE AUT CAN NOR NLD ISL AUS 0.5 ITA ESP PRT 0.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Non-BERD/GDP (%) 1. The regression line excludes Iceland, which is an important outlier. Source: OECD Main Science and Technology Indicators database and R&D database. 10

Figure 2.4 Non-business R&D intensity % of GDP, average per annum 1.25 1.00 0.75 0.50 0.25 0.00 Source: OECD Main Science and Technology Indicators database and R&D database. 11 ECO/CPE/WP1(2005)2/ANN1 Ireland UK Canada Switzerland Australia Japan Iceland 1996-2000 1981-86 2001

Figure 2.5 Business sector researchers % of total dependent employment, average per annum 1.0 0.8 0.6 0.4 0.2 0.0 Australia UK Ireland Canada Switzerland Iceland Japan 1996-2000 1981-85 2001 Source: OECD Main Science and Technology Indicators database and R&D database. 12

Figure 2.6. Composition of innovation spending 1 Community Innovation Survey countries, 1998-2000 intramural R&D machinery training extramural R&D external knowledge Panel A: Manufacturing versus services 2 Proportion of engagements in innovation spending 100% 80% 60% 40% 20% 0% manufacturing services Panel B: By firm size 3 Proportion of engagements in innovation spending 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% small firms medium firms large firms 1. The proportion of engagements in innovation spending is calculated as the ratio of the proportion of firms engaging in a specific type of innovation spending to the sum of the proportions of firms engaging in the various types of innovation spending. Firms can engage in multiple forms of spending. 2. The sectoral aggregates are calculated by taking a weighted average of the individual countries' observations for the sector and using as country weight the country's share in the total population of firms working in the sector in all included countries. 3. The size class aggregates are calculated by taking a weighted average of the individual countries' observations for the size class and using as country weight the country's share in the total population of firms working in the size class in all included countries. Source: Community Innovation Survey 3 (European Commission). 13

Figure 2.7 Total patents per million of working age population 1 Average per annum; residency of inventor 900 800 1996-2000 1986-90 2000 700 600 500 400 300 200 100 0 Ireland UK Iceland Canada Switzerland Japan (2) 1. Total patents are patent applications at the EPO for European countries, patent applications at the JPO for Japan, and patent grants (based on priority dates) at the USPTO for the United States and Canada. Australia is not included in the figure because comparable data on patent applications at the Australian patent office were not readily available. 2. As explained in Section 2.3 of the main text, the level of patent applications at the JPO is much larger because a different application has to be filed for each claim. The patent applications for Japan were divided by five to fit the scale of the figure. Their level is not comparable with the level of patent applications of other countries. Source: OECD Patent database; JPO and Maskus and McDaniel (1999) for data on Japan. 14

175 150 125 100 75 50 25 0 Source: OECD Patent database. Figure 2.8 Triadic patents per million of working age population Average per annum; residency of inventor 15 ECO/CPE/WP1(2005)2/ANN1 Ireland Australia Canada Iceland UK Japan Switzerland 1995-1999 1985-89 1999

0.3 0.2 0.2 0.1 0.1 0.0 Source: OECD Patent database. Figure 2.9 Triadic patents per million of business R&D spending 2000 PPP-adjusted R&D spending, average per annum 16 Ireland Canada Australia UK Iceland Japan Switzerland 1995-1999 1985-89 1999

Figure 2.10 Choice of protection method 1 Community Innovation Survey countries, 1998-2000 lead time secrecy complexity patent trademark design registration copyright Panel A: Manufacturing versus services 2 Proportion of aggregate protection 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% manuf services Panel B: By firm size 3 Proportion of aggregate protection 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% small firms medium firms large firms 1. The proportion of a specific type of protection in aggregate protection is calculated as the ratio of the proportion of firms using this specific type of protection to the sum of the proportions of firms using the various types of protection. Firms can use different forms of protection. 2. The sectoral aggregates are calculated by taking a weighted average of the individual countries' observations for the sector and using as country weight the country's share in the total population of firms working in the sector in all included countries. 3. The size class aggregates are calculated by taking a weighted average of the individual countries' observations for the size class and using as country weight the country's share in the total population of firms working in the size class in all included countries. Source: Community Innovation Survey 3 (European Commission). 17

Figure 2.11 Comparison of countries' innovative performance by various measures, 1998-2000 1 Successful innovators Engaging in intramural R&D 60 50 40 30 20 10 0 Greece United Kingdom Luxembourg Iceland Proportion of all firms (%) 1. Care has to be taken when interpreting cross-country comparisons made with the aggregated data in CIS, as there are differences in the sample size used in the respective national components of the survey. For,, and the United Kingdom, no data are available on the proportion of fims which engage in intramural R& Source: Community Innovation Survey 3 (European Commission). 18

Figure 2.12 Share of new products in turnover 1 1998-2000 Share in total turnover (%) 25 20 15 10 5 0 Iceland Luxembourg Greece 1. Care has to be taken when interpreting cross-country comparisons made with the aggregated data in CIS, as there are differences in the sample size used in the respective national components of the survey. Source: Community Innovation Survey 3 (European Commission). 19

Figure 2.13 Pure innovation and imitation 1 1998-2000 50 40 30 20 10 0 Greece United Kingdom Luxembourg Iceland true product innovation product imitation Proportion of all firms (%) 1. Care has to be taken when interpreting cross-country comparisons made with the aggregated data in CIS, as there are differences in the sample size used in the respective national components of the survey. Pure innovation refers to the proportion of firms which have introduced a product new to the market. Imitation refers to the proportion of firms which have introduced a product new to the enterprise but not new to the market. Data on the distinction between pure innovation and imitation are available only for product innovations, and not for process innovations. Source: Community Innovation Survey 3 (European Commission). 20

Figure 3.1 The state of tax and subsidisation policies Average per annum, 1996-2000 14 12 10 8 6 4 2 0 High tax incentives and high subsidies ESP (0.4) CAN (1.0) FRA (1.4) PRT (0.2) AUT (1.1) NLD (1.1) DNK (1.3) AUS (0.7) (1.9) IRE (0.9) NOR (0.9) GBR (1.2) DEU (1.6) SWE (2.8) BEL (1.4) FIN (2.0) ISL (1.0) CHE (1.9) JPN (2.1) ITA (0.5) R&D subsidies / Business R&D (per cent) High tax incentives and low subsidies Low tax incentives and high subsidies 0.65 0.7 0.75 0.8 0.85 0.9 0.95 1 1.05 1.1 1.15 Average Low tax incentives and low subsidies Average B-index 2 1. The numbers in parentheses are the average business R&D intensities in 1996-2000. 2. The B-index is defined as one minus the rate of tax subsidy for R&D. Source: OECD Main Science and Technology Indicators database for data on R&D subsidies; OECD, STI/EAS Division for data on the B-index. 21

Figure 3.2 Changes in tax and subsidisation policies Average per annum 1996-2000 versus Average per annum 1981-1985 10 5 0-5 -10-15 ESP DNK PRT NLD AUS IRE FRA AUT JPN FIN DEU NOR CHE ITA BEL CAN SWE Change in R&D subsidies /Business R&D (per cent) GBR -20-0.15-0.1-0.05 0 0.05 0.1 0.15 Change in the B-index1 1. The B-index is defined as one minus the rate of tax subsidy for R&D. Source: OECD Main Science and Technology Indicators database for data on R&D subsidies; OECD, STI/EAS Division for data on the B-index. 22

Figure 3.3 Business funding of non-business sector R&D 1 % points, average per annum 14.0 12.0 10.0 8.0 6.0 4.0 2.0 0.0 1. The last observed value is 2000 instead of 2001 for Australia, Ireland, and Switzerland; for, the last observed value is 1996. Source: OECD Main Science and Technology Indicators database. 23 ECO/CPE/WP1(2005)2/ANN1 Japan Australia Switzerland Canada Ireland UK Iceland 1996-2000 1981-85 2001

Figure 3.4 Convergence in intellectual property rights (Park index) 1985-2000 1.5 Absolute change in IPR 1985-2000 1.3 1.1 0.9 0.7 PRT CAN FIN IRE AUS SWE AUT ESP DEU NOR GBR 0.5 DNK 0.3 CHE ITA JPN R 2 = 0.566 0.1 FRA NLD BEL -0.1 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 IPR 1985 Source: Park and Wagh (2002) and Ginarte and Park (1997). 24

Figure 4.1 Financial factors Panel A: Financial development / GDP 1 Average 1996-2000, in per cent 400 350 300 250 200 150 100 50 0 Credit Equity financing Iceland Ireland Canada Australia Japan UK Switzerland Panel B: Profits / GDP Average 1996-2000, in per cent 40 35 30 25 20 15 10 5 0 Iceland Ireland Canada Australia Japan UK Switzerland 1. Financial development is defined as the sum of credit and equity financing. Credit refers to private credit by deposit money banks. Equity financing refers to the stock market capitalisation. Source: World Bank Financial Development and Structure database for data on financial development and OECD Analytical database for data on profits. 25

Figure 4.2 Evolution of the share of equity financing in financial development 1 Average per annum 1996-2000 versus Average per annum 1981-1985 80 70 60 50 SWE FIN DNK CHE NLD CAN GBR AUS Average 1996-2000 FRA IRE 40 ESP ITA NOR BEL JPN 30 ISL PRT DEU 20 10 AUT 0 0 10 20 30 40 50 60 70 80 Average 1981-1985 1. Financial development is defined as the sum of credit and equity financing. Equity financing refers to the stock market capitalisation. Source: World Bank Financial Development and Structure database. 26

Figure 4.3 PMR index 1 Panel A: Cross-country comparison in 1998 5 4 3 2 1 0 UK Australia Canada Japan Switzerland Ireland Panel B: Change between 1985 and 1998 UK Australia Source: OECD database. Canada Japan Switzerland Ireland -3.5-3.0-2.5-2.0-1.5-1.0-0.5 0.0 1. This OECD index of product market regulation covers 7 non-manufacturing sectors for which time-series data are available. Source: Nicoletti and Scarpetta (2003). 27

Figure 4.4 Index of FDI restrictions Panel A: Level in 1998 0.4 0.3 0.2 0.1 0 UK Ireland Switzerland Japan Australia Canada Panel B: Change between 1985 and 1998 UK Ireland Switzerland Japan Australia Canada -0.40-0.35-0.30-0.25-0.20-0.15-0.10-0.05 0.00 Source: Golub (2003). 28