http://www.smallbusinessinstitute.biz A B S T R A C T Keywords: Journal of Small Business Strategy 2021, Vol. 31, No. 01, 66-80 ISSN: 1081-8510 (Print) 2380-1751 (Online) ©Copyright 2021 Small Business Institute® w w w. j s b s . o rg Introduction VILNIUS TECH (Vilnius Gediminas Technical University), J. Basanavičiaus str. 28, 03224 Vilnius, Lithuania, ruta.baneliene@vilniustech.lt Industry impact on GDP growth in developed countries under R&D investment conditions Industry, GDP growth, OECD countries, R&D, Investment APA Citation Information: Banelienė, R. (2021). Industry impact on GDP growth in developed countries under R&D investment conditions. Journal of Small Business Strategy, 31(1), 66-80. The impact of industry on GDP growth is widely dis- cussed in light of the industrial revolution that arose in the mid-18th century as the first wave of innovation. Crafts and Mills (2017) provided a time-series analysis of annual es- timates of real GDP and industrial output covering 1270- 1913, the period before and after the first innovation wave. Their main findings were as follows: on average, the growth trend was 0.2% per year over the 500 years from 1270 to 1770; following the industrial revolution, the growth in real GDP per capita peaked at approximately 1.25% per year in the mid-nineteenth century. Among the conclusions of Solow (1957) based on the application of a crude production function to American data over the period 1909-1949 were that gross output per working hour doubled over the period, with 87.5% of the increase attributable to technical change and the remaining 12.5% to increased use of capital. Schumpeter’s (1934) theories and the concept of “cre- ative destruction” introduced the disruption of existing eco- nomic activity by innovations as a possibility for creating new ways of producing goods or services or entirely new industries (OECD/Eurostat. (2018). The strategic stimulus to economic development in Schumpeter’s (1934) analysis is innovation, defined as the commercial or industrial appli- cation of something new: a new product, process, or meth- od of production; a new market or source of supply; and a new form of commercial, business, or financial organi- zation. The innovational process incessantly revolutionizes the economic structure from within, incessantly destroying the old one and incessantly creating a new one (Schumpeter, 1934). Based on Schumpeter’s (1934) work, the Oslo Manual (2005) defined four types of innovations that encompass a wide range of changes in firms’ activities: product innova- tions, process innovations, organizational innovations and marketing innovations. These efforts were furthered in the 2018 Oslo Manual (OECD/Eurostat. (2018), and innova- tion was defined as a new or improved product or process that differs significantly from the unit’s previous products or processes and that has been made available to potential users (product) or brought into use by the unit (process). The impact of industry on GDP growth is widely discussed in light of the industrial revolution that arose as the first wave of innovations and has since been a common subject of theoretical and empirical research papers. However, the issue of R&D investment, industrial structure and the impact on GDP growth is still under discussion and requires much deeper investigation under conditions of global- ization. The results of this paper supported the hypotheses that growth of the industry share in gross value added has a higher impact on GDP growth in well-developed industrialized countries with high GDP per capita than in industrialized countries whose GDP per capita is at a lower level under business-financed R&D investment conditions. In addition, the multiplier effect of business-financed R&D investment and its impact on economic growth depend on the economic development level of a given industrialized country. The proposed hypotheses suggest that policy makers of less-developed countries should pay more attention to enhancing the quality of industry by introducing appropriate incentives rather than to increasing the share of industry in GDP, with a particular focus on the best practices of well-developed industrialized countries. Rūta Banelienė http://www.smallbusinessinstitute.biz http://www.jsbs.org 67 R. Banelienė Journal of Small Business Strategy / Vol. 31, No. 1 (2021) / 66-80 The impacts of innovations on firm performance range from effects on sales and market share to changes in productiv- ity and efficiency. Important impacts at the industry and national levels are changes in international competitiveness and in total factor productivity, knowledge spillovers from firm-level innovations, and an increase in the amount of knowledge flowing through networks. According to the Oslo Manual (2018), many factors could have an impact on boosting innovations in the busi- ness sector, including internal factors, such as general re- sources of the firm, management capabilities, workforce skills and human resources management, and technological capabilities; and external factors, such as knowledge flows, including the diffusion of innovations and the collection of data on knowledge flows, the location of business activities, markets and environment for business innovation, public policy in the forms of regulations and government support programs, public infrastructure, the macroeconomic policy environment, and the social and natural environment for in- novation. The revived interest in industrial policy comes at the time of a new technological transformation and accompa- nies the arrival of radical and disruptive technologies asso- ciated with the applications of artificial intelligence, auto- mation and machine learning. Industry 4.0 and the so-called “Fourth Industrial Revolution” embody technologies such as advanced robotics, increased automation, digitalization and data exchange in manufacturing technologies supported by artificial intelligence, cyber-physical systems, platform economy innovations and cloud computing (Bailey, 2019). Comparable and representative data for 2015 on the deployment of industrial robot technologies show that Ko- rea and Japan lead in terms of robot intensity (i.e., the indus- trial stock of robots over manufacturing value added). The robot intensity in these economies is approximately three times that in the average OECD country (OECD, 2017). Innovative effort is on the rise as a share of economic activity. Investment in knowledge has grown more rapidly than investment in machinery and equipment since the mid- 1990s in most OECD countries (OECD, 2007). There is no doubt that innovation is fundamental to economic development and growth. Innovation enables not only firms but also industries and even countries to com- pete with each other (Slaper & Justis, 2010). Innovation is the engine of economic growth: innovation and its diffusion within and across national boundaries rather than the accu- mulation of capital explain the economic growth rates of countries (Sainsbury, 2020). The novelty of this study is its examination of the link between industry and GDP growth in developed countries under business R&D investment conditions, including the price of capital. Previous research and the majority of exist- ing models investigated the relation between GDP growth and industry without keeping in mind R&D investment con- ditions, as will be shown in the section “Theoretical Ap- proach and Empirical Evidence” of this paper. The purpose of this study is to analyze the impact of industry and business R&D investment on GDP growth in OECD countries. Seeking to examine the hypothesis on the relation of industry and GDP growth in developed coun- tries under business R&D investment conditions, data for 36 OECD countries were used, and regression analysis was applied. The structure of this article is as follows: the the- oretical approach and empirical evidence clarify the direc- tion for the empirical study and model development, which is a proven novelty of this study in comparison to existing empirical evidence. The empirical background part ana- lyzes OECD countries’ statistical data with a special focus on industry, its structure and country segmentation by GDP per capita. The part regarding the methodology, model and data describes the developed model for testing the proposed hypotheses, and the results part provides the modeling out- comes. The conclusion and discussion section summarizes the major insights of this study and highlights the implica- tions for policy makers. Theoretical Approach and Empirical Evidence The links between technology and the economy were at the very heart of the Industrial Revolution, and the suc- cessful conversion of many inventions into profitable inno- vations in numerous small but growing firms made possible the acceleration of productivity growth in the leading sec- tors; this was not a linear process any more than it is in the present (Freeman, 2019). Fagerberg & Verspagen (2020), by following Free- man’s analytical framework on long-run economic devel- opment as a succession of technological regimes with quite different properties, stressed that at the core of each regime is a constellation of radical innovations, the diffusion of which generates many new applications and, for a while, strong economic growth. The relationship between innovation and economic change is conceptualized as a two-way link. On the one hand, the former is indeed a major driver of economic per- formance; on the other hand, specific economic and busi- ness conditions can explain the way particular innovations emerge, for example, when exports and profits shape in- 68 R. Banelienė Journal of Small Business Strategy / Vol. 31, No. 1 (2021) / 66-80 dustries’ ability to introduce product innovations (Pianta, 2017). The results for manufacturing alone are somewhat dif- ferent. Although R&D maintains a strong influence on in- novative sales, technology adoption loses its significance; manufacturing industries have been characterized by wide- spread processes of restructuring, and technology adoption from external sources (particularly regarding new machin- ery) has mainly had the effect of reducing labor use, rath- er than increasing innovative sales (Bogliacino & Pianta, 2012). Denoncourt (2020), with reference to Rahman et al. (2017), noted that society has changed, with the decline of traditional manufacturing and simple supply chains and their frequent replacement with digital and intellectual property (IP)-reliant business models operating in the intan- gible (virtual or weightless) economy. Firms with new digi- tal technologies create new markets and value networks that impact established markets, firms, products and alliances. Artificial intelligence (AI) radically shortens the time needed to discover new industrial materials, sometimes from years to days (Chen, 2017). Currently, the greatest AI commercial potential for advanced manufacturing is expect- ed to exist in supply chains, logistics and process optimiza- tion (Chui et al., 2018). In addition to having direct uses in production, AI used in logistics enables real-time fleet management while significantly reducing fuel consumption and other costs (OECD, 2018). Ulku (2007) provided an empirical analysis of the re- lationship between R&D intensity, rate of innovation and growth rate of output in four manufacturing sectors, namely, chemicals, drugs and medicine, electrical and electronics, and machinery and transport, in 17 OECD countries over the period 1981-1997. Her findings suggest that knowledge stock is the main determinant of innovation in all four man- ufacturing sectors and that R&D intensity increases the rate of innovation in the chemicals, electrical and electronics, and drugs and medicine sectors. In addition, the rate of in- novation has a positive effect on the growth rate of output in all sectors. Coccia (2009) analyzed the relationship between pro- ductivity growth and R&D investment. The econometric analysis showed that gross domestic expenditure on R&D (GERD), as a percentage of GDP, is an important driver of productivity growth. The optimization showed that the long-run magnitude of GERD between 2.3% and 2.5% maximizes productivity growth. He stated that countries should focus their targets on R&D investment in the amount of approximately 2.3-2.5% of GDP in the long run. Howev- er, this problem is still under discussion and requires much deeper investigation through the examination of other in- dicators, including GDP per capita. The data collected for this study could not clearly prove or reject Coccia’s hypoth- esis or indicate whether there is an optimum level of R&D investment, based on the relation between business R&D expenditure and GDP per capita. Frick et al. (2019) argued that the differences in R&D intensity observed among subsectors of the manufacturing sector are primarily a result of differing returns to innovation and differing innovation behaviors. According to Guellec et al. (2001), R&D is important for productivity and economic growth. Business R&D has high spillover effects; it enhanc- es the ability of the business sector to absorb technology coming from abroad or from government and university re- search. The social return on business R&D is then higher than its private return, which is a possible justification for government support of business R&D. Using the innovation index and its component indexes as a measure of the innovative environment prevailing in US states, Cheung (2014) found that the more innovative a state is, the higher its per capita real GDP and per capita personal income are. Higher per capita personal income is associated with both the availability of human capital for innovative activities and the presence of the economic dy- namics that facilitate those activities. Using data on top R&D spenders in the US and the EU from 2004 until 2012, Castellani et al. (2016) investigated the sources of the US/EU productivity gap. They found ro- bust evidence that US firms have a higher capacity to trans- late R&D into productivity gains (especially in high-tech industries), and this contributes to explaining the higher productivity of US firms. Conversely, EU firms are more likely to achieve productivity gains through capital-embod- ied technological change, at least in medium- and low-tech sectors. Sainsbury (2020) pointed out that the growth of high value-added per capita industries is the key to economic development, and the creation of competitive advantage and production efficiency by innovation is the way firms create high value-added per capita. Governments need to draw up policies that support industry in four main areas: firm financing and governance, regional policy, the national system of education and training, and the national system of innovation. Guimón & Paunov (2019) pointed out that three fac- tors are particularly important for evaluating a country’s policy mix: 1) the characteristics of the business sector, 2) the characteristics of higher education institutions and pub- 69 R. Banelienė Journal of Small Business Strategy / Vol. 31, No. 1 (2021) / 66-80 lic research institutes and of their research, and 3) macro- economic conditions. Appelt et al. (2019), based on the OECD R&D Tax In- centives Database for 36 OECD countries and partner econ- omies, found that over the past two decades, many countries have increased the availability, simplicity of use and gener- osity of R&D tax incentives. Empirical Background Seeking to investigate industry directions that have a high impact on economic growth, this analysis is based on the data of OECD countries. In the first step, an overview of all 36 OECD countries was made considering two criteria – manufacturing share in gross value added and GDP per capita. In the second step, we analyzed the data in more detail and found that 17 OECD countries could be called indus- trialized countries with high GDP per capita. The largest selection of countries – 13 countries whose GDP per cap- ita is higher than $40,000 and whose manufacturing share in gross value added is 10-19.9% – included Austria, Bel- gium, Canada, Denmark, Finland, France, Iceland, Italy, Netherlands, New Zealand, Sweden, the United Kingdom, and the United States. For further analysis, four countries from other segments were included: Germany and Japan, where manufacturing share in gross value added exceeds 20% and whose GDP per capita higher than $40,000; Swit- zerland, where manufacturing share is in between 10-19.9% and GDP per capita is above $60,000; and Ireland, whose manufacturing share in gross value added exceeds 30% and whose GDP per capita is above $60,000 (see Table 1). Furthermore, the industry structure in 16 selected coun- tries by the top three industries was analyzed in more detail. Differences in terms of population were accounted for, and population figures were additionally included in this step of analysis. Eurostat data for the year 2017 were used for Euro- pean Union countries (Germany, Sweden and United King- dom data for 2016) and Iceland; and 2016 data were used for Switzerland. In addition, national statistical offices data were used for 2017 for the US, Japan, New Zealand, and Canada (Bureau of Economic Analysis 2019; Statistics of Japan (2019); Stats NZ (2019); & Statistics Canada (2019), respectively). Population data for 2018 were extracted from the World Bank (2019) database. Ireland was excluded from this step of the analysis due to the unavailability of detailed Table 1 OECD countries by GDP per capita in relation to manufacturing share in gross added value, 2017 GDP/Capita Manufacturing Share in Gross Added Value Less than 10% 10-19.9% 20-29.9% 30% and More Less than $20,000 Mexico $20,000-$39,999 Greece, Poland, Portugal, Spain, Turkey, Chile, Es- tonia, Israel, Latvia, Lithuania Czech Republic, Hungary, Slovakia, Slovenia Korea $40,000-$59,999 Australia Austria, Belgium, Canada, Denmark, Finland, France, Ice- land, Italy, Nether- lands, New Zealand, Sweden, United Kingdom, United States Germany, Japan $60,000 and More Luxembourg, Norway Switzerland Ireland Source: Prepared by author, based on OECD data (OECD 2019b, 2029d). Data on manufacturing share in gross value added for Iceland and New Zealand for 2016 and for Canada for 2015. 70 R. Banelienė Journal of Small Business Strategy / Vol. 31, No. 1 (2021) / 66-80 data on the manufacturing structure (Eurostat, 2019b). After data analysis, it was found that most of the select- ed countries had the manufacture of food products, bever- ages and tobacco products as one of the top 3 industries (13 of 16 countries; of which 5 small countries had a population under 10 million). This suggests that high-income countries should pay attention to the food industry and seek to devel- op a sustainable food supply with appropriate quality stan- dards for their populations. The second most common top industry is machinery and equipment manufacturing: more than half of the selected countries (10 of 16), both small and large developed countries, have this industry as one of the top 3 manufacturing sectors. Basic metal and fabricat- ed metal product manufacturing is in third place: 8 of the 16 selected countries have developed this industry, and the importance of this industry likewise does not depend on a country’s size in terms of population. Motor vehicles, trailers, semitrailers and other trans- port equipment manufacturing and chemicals and chem- ical product manufacturing are industry priorities in large countries with a population above 10 million. This could be explained by historical aspects based on the first wave of innovations and the owned natural resources that allowed the development of the transport equipment manufacturing sector in Canada, France, Sweden, the United Kingdom, Germany and Japan, even though the extraction of metal ores is currently shrinking in many of these countries (Ma- terials Flow Analysis Portal, 2019). The priority of chem- ical product manufacturing in Belgium, the Netherlands and United States could be explained by the same historical reason – owned natural resources and inventions. Although this industry is not found only in large countries, there is a great difference in population between the US and Belgium and the Netherlands. Basic pharmaceutical product and pharmaceutical preparation manufacturing and computer, electronic and op- tical product manufacturing are the top 3 priority industries in the selected countries, both small and large countries. In addition, this research step showed that small, well-developed industrialized countries have used their natural resources for boosting economic growth, and this is one of their top 3 manufacturing directions: for example, Finland has used wood as its own natural resource for de- veloping wood, paper, printing and reproduction manufac- turing, and New Zealand has developed petroleum and coal product manufacturing. However, industries based purely on countries’ own natural resources cannot be considered a trend for all selected OECD industrialized countries and were excluded from further analysis in this study. As the first-stage outcome of the data analysis, seven major manufacturing directions could be observed: 1) food products, beverages and tobacco products; 2) machinery and equipment; 3) basic metals and fabricated metal products; 4) motor vehicles, trailers, semitrailers and other transport equipment; 5) chemicals and chemical products; 6) basic pharmaceutical products and pharmaceutical preparations; and 7) computer, electronic and optical products, which could create high value added in the industrial sector, as proven by their development as top directions in OECD in- dustrialized countries whose GDP per capita is higher than $40,000. In addition, it should be noted that only two of the seven directions – motor vehicles, trailers, semitrailers and other transport equipment manufacturing and chemicals and chemical products manufacturing – belong to the top 3 industries of large countries, whose population is above 10 million. Eurostat (2019a) classified only two of the seven high- lighted manufacturing sectors as high-technology sectors: manufacture of basic pharmaceutical products and phar- maceutical preparations and manufacture of computer, electronic and optical products. According to technological intensity, three of the seven manufacturing industries are considered medium-high technology: manufacture of chem- icals and chemical products, manufacture of machinery and equipment, and manufacture of motor vehicles, trailers and semitrailers, and other transport equipment. Basic metal and fabricated metal product manufacturing is classified as a medium-low-technology industry, and food product, bev- erage and tobacco product manufacturing is classified as a low-technology industry. A similar classification is provid- ed by the OECD (2011). Innovation in low- and medium-technology industries (LMTs) often receives less attention than innovation in high-technology industries. However, innovation in LMTs can have a substantial impact on economic growth, owing to the weight of these sectors in the economy (OECD/Eu- rostat, 2005). Method Regression analysis was used for modeling based on the following equation: 𝐺𝐺𝐺𝐺𝐺𝐺 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 = 𝑐𝑐 + 𝐵𝐵𝐵𝐵𝐵𝐵&𝐺𝐺 𝐺𝐺𝐺𝐺𝐺𝐺 + 𝐼𝐼𝐼𝐼𝐺𝐺 𝐺𝐺𝐺𝐺𝐺𝐺 + 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿 + 𝜀𝜀 (1) 71 R. Banelienė Journal of Small Business Strategy / Vol. 31, No. 1 (2021) / 66-80 Here, BER&D/GDP refers to the business-financed gross domestic expenditure on R&D as a percentage of GDP, IND/GAV denotes the share of industry as a percent- age of gross value added, and LTIR denotes the long-term interest rate. This study uses OECD data from 36 countries (see Ap- pendix). The model hypotheses (HP) are as follows: HP1. Growth in the share of industry has a higher impact on GDP growth in well-developed industrialized countries whose GDP per capita is higher than $40,000 than in indus- trialized countries where GDP per capita is at a lower level under business-financed R&D investment conditions. HP2. The multiplier effect of business-financed R&D in- vestment and its impact on economic growth depend on the industrialized country’s economic development level. Results The panel least squares method was applied for the estimations. Annual data for 36 OECD countries over the period 2014-2017 were used for modeling. Three countries were excluded due to lack of data on the long-term inter- est rate (Estonia, Turkey) and on business-financed gross domestic expenditure on R&D (Australia). Modeling was performed using the Eviews10 program. The estimation results prove the hypothesis that the growth of industry share in gross value added has a high- er impact on GDP growth in well-developed industrialized countries whose GDP per capita is higher than $40,000 than in industrialized countries whose GDP per capita is at a lower level under business-financed R&D investment con- ditions. For the baseline estimation, data were used for 33 OECD countries, excluding 3 OECD countries due to the lack of data for Australia on business sector R&D expen- diture and for Estonia and Turkey on the long-term interest rate. The estimation covers the period 2014-2017 and in- cludes 105 (total panel unbalanced) observations. The pan- el least squares method with white cross-section standard errors and covariance and fixed cross-sectional variables (dummy variables) was used for the baseline estimation as shown in Equation (2). R-squared (R2) = 0.9910; adjusted R-squared (R2) = 0.9864; D-W = 1.7953. The estimation shows a positive impact of R&D in- vestment by business sector on GDP per capita, where one additional percentage point of R&D investment growth rais- es GDP per capita by $8,569. In addition, an additional per- centage point of industry share growth in gross added value raises GDP per capita by $1,175. A one percentage point increase in the long-term interest rate has a negative impact, leading GDP per capita to drop by $1,314 (see Equation (2) and Table 2). In the second modeling step, the same estimation was made for 17 industrial OECD countries whose industry share in gross value added is above 10% and whose GDP per capita is above $40,000, namely, Austria, Belgium, Canada, Denmark, Finland, France, Iceland, Italy, Neth- erlands, New Zealand, Sweden, the United Kingdom, the United States, Germany, Japan, Switzerland and Ireland (see Table 1). The data covers the same period, 2014-2017, and includes 49 (total panel unbalanced) observations. The panel least squares method with white cross-section stan- dard errors and covariance and fixed cross-section variables (dummy variables) was used for this estimation. R-squared (R2) = 0.9763; adjusted R-squared (R2) = 0.9608; D-W = 1.7410. The second estimation results show the same relations as the baseline estimation: positive relationships between GDP per capita and business sector R&D investment, as well as between GDP per capita and industry share in gross value added, and a negative relation with the long-term in- terest rate. However, in the second estimation, the impact of busi- ness-financed R&D investment is 1.54 times higher than that in the estimation for all selected 33 OECD countries. Additionally, industry share growth in gross value added has a higher impact on GDP per capita growth but only of 6.4% in comparison to the baseline estimation results. The long-term interest rate in this case is 1.45 times more vul- nerable than the estimation output for all 33 selected OECD countries (see Equation (3) and Table 3). 𝐺𝐺𝐺𝐺𝐺𝐺 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 = 11334.92 + 13223.27 𝐵𝐵𝐵𝐵𝐵𝐵&𝐺𝐺 𝐺𝐺𝐺𝐺𝐺𝐺 + 1251.10 𝐼𝐼𝐼𝐼𝐺𝐺 𝐺𝐺𝐺𝐺𝐺𝐺 − 1912.08𝐿𝐿𝐿𝐿𝐼𝐼𝐵𝐵 + 𝜀𝜀 (3) 𝐺𝐺𝐺𝐺𝐺𝐺 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 = 13507.80 + 8568.59 𝐵𝐵𝐵𝐵𝐵𝐵&𝐺𝐺 𝐺𝐺𝐺𝐺𝐺𝐺 + 1175.38 𝐼𝐼𝐼𝐼𝐺𝐺 𝐺𝐺𝐺𝐺𝐺𝐺 − 1314.23𝐿𝐿𝐿𝐿𝐼𝐼𝐵𝐵 + 𝜀𝜀 _ _ 72 R. Banelienė Journal of Small Business Strategy / Vol. 31, No. 1 (2021) / 66-80 The third estimation was made for the 14 selected OECD industrialized countries whose GDP per capita was below $40,000 and whose industry share in gross value added was higher than 10%: Mexico, Greece, Poland, Por- tugal, Spain, Chile, Israel, Latvia, Lithuania, Czech Repub- lic, Hungary, Slovakia, Slovenia and Korea (see Table 1). Turkey and Estonia were excluded from the estimation due to lack of data on the long-term interest rate. The data cov- er the same period, 2014-2017, and include 52 (total panel unbalanced) observations. The panel least squares method with fixed cross-section variables (dummy variables) was used for the estimation. Table 2 Review regression estimation: All selected OECD countries Variable Coefficient Std. Error t-Statistic Prob. C 13507.80 1961.109 6.887837 0.0000 BER&D/GDP 8568.587 1745.701 4.908393 0.0000 IND/GAV 1175.383 38.29223 30.69509 0.0000 LTIR -1314.225 435.3570 -3.018730 0.0036 Effects Specification Cross-Section Fixed (Dummy Variables) R-squared 0.990959 Mean dependent var 40001.53 Adjusted R-squared 0.986373 SD dependent var 13418.09 SE of regression 1566.331 Akaike info criterion 17.81672 Sum squared residuals 1.69E+08 Schwarz criterion 18.72665 Log likelihood -899.3778 Hannan-Quinn crit. 18.18544 F-statistic 216.0902 Durbin-Watson stat 1.795279 Prob(F-statistic) 0.000000 Table 3 Review regression estimation: 17 selected OECD industrialized countries with high GDP per capita Variable Coefficient Std. Error t-Statistic Prob. C 11334.92 4172.468 2.716598 0.0110 BER&D/GDP 13223.27 2977.429 4.441171 0.0001 IND/GAV 1251.104 59.92816 20.87673 0.0000 LTIR -1912.082 163.0936 -11.72383 0.0000 Effects Specification Cross-Section Fixed (Dummy Variables) R-squared 0.976320 Mean dependent var 48442.13 Adjusted R-squared 0.960805 SD dependent var 8102.611 SE of regression 1604.129 Akaike info criterion 17.89035 Sum squared residuals 74623633 Schwarz criterion 18.66252 Log likelihood -418.3136 Hannan-Quinn crit. 18.18331 F-statistic 62.92906 Durbin-Watson stat 1.741032 Prob(F-statistic) 0.000000 73 R. Banelienė Journal of Small Business Strategy / Vol. 31, No. 1 (2021) / 66-80 This estimation shows that the industry share in gross value added is irrelevant in modeling the impact on GDP per capita by Equation (1) (p = 0.93). After elimination of the industry indicator from the equation, the estimation shows the same positive relation between GDP per capita and R&D investment by busi- ness sector and a negative relation to the long-term inter- est rate, as in the two previous estimations. The level of GDP per capita at which a one percentage point increase in business-financed R&D investment could enhance GDP per capita growth by $7,647 is more than twice as high ($27,511) as the level in the baseline and second estima- tions: for comparison, in the baseline estimation for all se- lected OECD countries, the GDP per capita starting point is $13,508, while in the second estimation for industrialized countries with high GDP per capita, it is $11,335. In addi- tion, the impact of R&D investment by the business sector is lower (in the baseline estimation, it is equal to $8,569, and in the second estimation, it is equal to $13,223) (see equations (2), (3) and (4)). This supports the hypothesis that the multiplier effect of business-financed R&D investment and its impact on economic growth depend on the economic development level of a given industrialized country. 𝐺𝐺𝐺𝐺𝐺𝐺 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 = 27510.85 + 7647.12 𝐵𝐵𝐵𝐵𝐵𝐵&𝐺𝐺 𝐺𝐺𝐺𝐺𝐺𝐺 − 1213.18𝐿𝐿𝐿𝐿𝐿𝐿𝐵𝐵 + 𝜀𝜀 (4) R-squared (R2) = 0.9536; adjusted R-squared (R2) = 0.9343; D-W = 1.5301. In the third estimation, growth of the long-term interest rate has a negative impact on economic growth, although it has a lower negative impact than it does in the baseline ($1,314) and second ($1,912) estimations: a one percent- age point higher interest rate will shrink GDP per capita by $1,213 (see Equations (2), (3) and (4)). The estimations support the hypotheses that growth in the industry share in gross value added has a higher impact on GDP growth in well-developed industrialized countries where GDP per capita is higher (in this case, more than $40,000) than in industrialized countries whose GDP per capita is at a lower level under business-financed R&D Table 4 Review regression estimation: industrial OECD countries with low GDP per capita Variable Coefficient Std. Error t-Statistic Prob. C 27510.85 2429.926 11.32168 0.0000 BER&D/GDP 7647.120 3064.581 2.495323 0.0173 LTIR -1213.175 271.2666 -4.472262 0.0001 Effects Specification Cross-Section Fixed (Dummy Variables) R-squared 0.953621 Mean dependent var 29684.26 Adjusted R-squared 0.934296 SD dependent var 5217.581 SE of regression 1337.408 Akaike info criterion 17.48251 Sum squared residuals 64391738 Schwarz criterion 18.08290 Log likelihood -438.5454 Hannan-Quinn crit. 17.71269 F-statistic 49.34753 Durbin-Watson stat 1.530123 Prob(F-statistic) 0.000000 investment conditions. In addition, the multiplier effect of business-financed R&D investment and its impact on eco- nomic growth depend on the economic development level of the industrialized country. Therefore, policy makers of less-developed countries should pay more attention to the rising quality of industry by introducing appropriate incentives, for example, R&D tax incentives, than to the share of industry in GDP, with a particular focus on the best practices of well-developed industrialized countries. Moreover, industrial companies should focus on increasing their efficiency by implementing the best practices of well-developed countries’ industrial companies. _ 74 R. Banelienė Journal of Small Business Strategy / Vol. 31, No. 1 (2021) / 66-80 Discussion and Conclusion The problem regarding the optimum level of R&D in- vestment is still under discussion and requires much deeper investigation to examine its relation not only to productivity but also to other indicators, including GDP per capita. The data collected for this research could not clearly prove or deny the Coccia (2009) hypothesis or indicate whether there is an optimum level of R&D investment in the long run. This study highlighted that seven major manufactur- ing directions can create high value added in the industri- al sector, as proven by their development as top industries in industrialized OECD countries whose GDP per capita is higher than $40,000. Although only two of the seven manu- facturing sectors belong to the high-technology sector, three are medium-high-technology industries, one is classified as a medium-low-technology industry, and one is classified as a low-technology industry. However, innovation in low- and medium-technology industries can have a substantial impact on economic growth, owing to the weight of these sectors in the economy. The modeling and estimations supported the hypoth- eses that growth of the industry share in gross value add- ed has a higher impact on GDP growth in well-developed industrialized countries whose GDP per capita is high (in this case, more than $40,000) than in industrialized coun- tries whose GDP per capita is at a lower level under busi- ness-financed R&D investment conditions. In addition, the multiplier effect of business-financed R&D investment and its impact on economic growth depend on the economic de- velopment level of a given industrialized country. Policy makers in less-developed industrialized coun- tries should pay more attention to increasing the quality of industry by introducing appropriate incentives than to en- hancing the industry share in GDP, with a particular focus on the best practices of well-developed industrialized coun- tries. The research limitations are the limited scope of the variables used, which could have an impact on the relation between industry and GDP growth in developed countries under business R&D investment conditions. Another lim- itation concerns the lack of investigation of the industrial structures of OECD industrialized countries with a lower GDP per capita. However, the estimations show that even if the manufacturing structures of countries with lower GDP per capita fully or partially reflect the manufacturing struc- tures of well-developed industrial OECD economies, their industries do not have the same impact on economic growth. Therefore, OECD industrialized countries with lower GDP should make efforts to improve their manufacturing sectors by enhancing their quality, changing their supply chains, focusing on high added value activities and transferring the other activities to less-developed countries. References Appelt, S., Galindo-Rueda, F., & González Cabral, A. C. (2019). Measuring R&D tax support: Findings from the new OECD R&D tax incentives database. (OECD Science, Technology and Industry Working Papers No. 6). https://doi.org/10.1787/d16e6072-en Bailey, D., Glasmeier, A., Tomlinson, P. R., & Tyler, P. (2019). Industrial policy: New technologies and trans- formative innovation policies?. Cambridge Journal of Regions, Economy and Society, 12(2), 169–177. https://doi.org/10.1093/cjres/rsz006 Bogliacino, F., & Pianta, M. (2012). Profits, R&D, and in- novation—A model and a test. Industrial and Corpo- rate Change, 22(3), 649–678. https://doi.org/10.1093/ icc/dts028 Bureau of Economic Analysis. (2019). United States’ value added by industry data. BEA. https://apps.bea.gov/ iTable/index_industry_gdpIndy.cfm Castellani, D., Piva, M., Schubert, T., & Vivarelli, M. (2016). R&D and productivity in the US and the EU: Sectoral specificities and differences in the crisis. Technological Forecasting and Social Change 138, 279-291. Chen, S. (2017). The AI Company that helps Boeing cook new metals for jets. Wired. https://www.wired.com/ story/the-ai-company-that-helps-boeing-cook-new- metals-for-jets Cheung, O. L. (2014). Impact of innovative environment on economic growth: An examination of state per cap- ita GDP and personal income. Journal of Business & Economics Research, 12(3), 257-270. https://doi. org/10.19030/jber.v12i3.8730 Chui, M., Manyika, J., Miremadi, M., Henke, N., Chung, R., Nel, P., & Malhotra, S. (2018). Notes from the AI frontier: Insights from hundreds of use cases. Mck- insey Global Institute. https://www.mckinsey.com/~/ media/mckinsey/featured%20insights/artificial%20 intelligence/notes%20from%20the%20ai%20fron- tier%20applications%20and%20value%20of%20 deep%20learning/notes-from-the-ai-frontier-insights- from-hundreds-of-use-cases-discussion-paper.ashx Coccia, M. (2009). What is the optimal rate of R&D invest- ment to maximize productivity growth?. Technologi- cal Forecasting and Social Change, 76(3), 433-446. https://doi.org/10.1016/j.techfore.2008.02.008 Crafts, N., & Mills, T. C. (2017). Six centuries of British https://doi.org/10.1787/d16e6072-en https://doi.org/10.1093/cjres/rsz006 https://doi.org/10.1093/icc/dts028 https://doi.org/10.1093/icc/dts028 https://www.wired.com/story/the-ai-company-that-helps-boeing-cook-new-metals-for-jets https://www.wired.com/story/the-ai-company-that-helps-boeing-cook-new-metals-for-jets https://www.wired.com/story/the-ai-company-that-helps-boeing-cook-new-metals-for-jets https://doi.org/10.19030/jber.v12i3.8730 https://doi.org/10.19030/jber.v12i3.8730 https://www.mckinsey.com/~/media/mckinsey/featured%20insights/artificial%20intelligence/notes%20from%20the%20ai%20frontier%20applications%20and%20value%20of%20deep%20learning/notes-from-the-ai-frontier-insights-from-hundreds-of-use-cases-discussion-paper.ashx https://www.mckinsey.com/~/media/mckinsey/featured%20insights/artificial%20intelligence/notes%20from%20the%20ai%20frontier%20applications%20and%20value%20of%20deep%20learning/notes-from-the-ai-frontier-insights-from-hundreds-of-use-cases-discussion-paper.ashx https://www.mckinsey.com/~/media/mckinsey/featured%20insights/artificial%20intelligence/notes%20from%20the%20ai%20frontier%20applications%20and%20value%20of%20deep%20learning/notes-from-the-ai-frontier-insights-from-hundreds-of-use-cases-discussion-paper.ashx https://www.mckinsey.com/~/media/mckinsey/featured%20insights/artificial%20intelligence/notes%20from%20the%20ai%20frontier%20applications%20and%20value%20of%20deep%20learning/notes-from-the-ai-frontier-insights-from-hundreds-of-use-cases-discussion-paper.ashx https://www.mckinsey.com/~/media/mckinsey/featured%20insights/artificial%20intelligence/notes%20from%20the%20ai%20frontier%20applications%20and%20value%20of%20deep%20learning/notes-from-the-ai-frontier-insights-from-hundreds-of-use-cases-discussion-paper.ashx https://www.mckinsey.com/~/media/mckinsey/featured%20insights/artificial%20intelligence/notes%20from%20the%20ai%20frontier%20applications%20and%20value%20of%20deep%20learning/notes-from-the-ai-frontier-insights-from-hundreds-of-use-cases-discussion-paper.ashx https://doi.org/10.1016/j.techfore.2008.02.008 75 R. Banelienė Journal of Small Business Strategy / Vol. 31, No. 1 (2021) / 66-80 economic growth: A time-series perspective. European Review of Economic History, 21, 141–158. https://doi. org/10.1093/ereh/hew020 Denoncourt, J. (2020). Companies and UN 2030 sustain- able development goal 9 industry, innovation and in- frastructure. Journal of Corporate Law Studies, 20(1), 199-235. https://doi.org/10.1080/14735970.2019.1652 027 Eurostat. (2019a). Glossary: High-tech classification of manufacturing industries. Eurostat. https://ec.europa. eu/eurostat/statistics-explained/index.php/Glossa- ry:High-tech_classification_of_manufacturing_indus- tries Eurostat. (2019b). National accounts aggregates by indus- try data. Eurostat. https://appsso.eurostat.ec.europa.eu/ nui/show.do?dataset=nama_10_a64&lang=en Fagerberg, J., & Verspagen, B. (2020). Innovation–diffu- sion, the economy and contemporary challenges: A comment. Industrial and Corporate Change, 29(4), 1067–1073. https://doi.org/10.1093/icc/dtaa019 Freeman, C. (2019). History, co-evolution and econom- ic growth. Industrial and Corporate Change, 28(1), 1–44. https://doi.org/10.1093/icc/dty075 Frick, F., Jantke, C., & Sauer J. (2019). Innovation and pro- ductivity in the food vs. the high-tech manufacturing sector. Economics of Innovation and New Technology, 28(7), 674-694. https://doi.org/10.1080/10438599.201 8.1557405 Guellec, D., & Van Pottelsberghe de la Potterie, B. (2001). R&D and productivity growth: Panel data analysis of 16 OECD countries. OECD Economic Studies, 33(II), 103-126. Guimón, J., & Paunov, C. (2019). Science-industry knowledge exchange: A mapping of policy instru- ments and their interactions. OECD Science, Tech- nology and Industry Policy Papers, 66. https://doi. org/10.1787/66a3bd38-en Materials Flow Analysis Portal. (2019). Domestic ex- traction of Canada, France, Sweden, United Kingdom, Germany and Japan in 1970-2017. http://www.materi- alflows.net/ OECD. (2007). Innovation and growth: Rationale for an innovation strategy. OECD. https://www.oecd.org/sci- ence/inno/39374789.pdf OECD. (2011). ISIC Rev. 3 technology intensity definition: Classification of manufacturing industries into cate- gories based on R&D intensities. OECD Directorate for Science, Technology and Industry Economic Anal- ysis and Statistics Division. https://www.oecd.org/sti/ ind/48350231.pdf OECD. (2017). OECD Science, Technology and Industry Scoreboard 2017: The digital transformation. OECD. OECD. (2018). OECD Science, Technology and Innovation Outlook 2018: Adapting to technological and societal disruption. OECD. OECD. (2019a). Business-financed GERD as a percentage of GDP data. OECD Stat. https://stats.oecd.org/Index. aspx?DataSetCode=MSTI_PUB OECD. (2019b). Gross domestic product (GDP) per capita data. OECD.Stat https://data.oecd.org/gdp/gross-do- mestic-product-gdp.htm OECD. (2019c). Long-term interest rate data. OECD Data. https://data.oecd.org/interest/long-term-interest-rates. htm OECD. (2019d). Value added by activity data. OECD Data. https://data.oecd.org/natincome/value-added-by-activ- ity.htm#indicator-chart OECD/Eurostat. (2005). Oslo manual: Guidelines for col- lecting and interpreting innovation data. (3rd ed.), OECD. OECD/Eurostat. (2018). Oslo manual 2018: Guidelines for collecting, reporting and using data on innovation. (4th ed.).OECD. https://doi.org/10.1787/9789264304604- en Pianta, M. (2017). Innovation and economic change. Eco- nomics of Innovation and New Technology, 26(8), 683- 688. https://doi.org/10.1080/10438599.2016.1257447 Rahman, A. A., Hamid, U. Z. A., & Chin, T. A. (2017). Emerging technologies with disruptive effects: A re- view. Perintis eJournal, 7(2), 111-128. Sainsbury, D. (2020). Toward a dynamic capability the- ory of economic growth. Industrial and Corporate Change, 29(4), 1047–1065. https://doi.org/10.1093/ icc/dtz054 Schumpeter, J. A. (1934). The theory of economic develop- ment: An inquiry into profits, capital, credit, interest and the business cycle. Harvard University Press. Slaper, T., & Justis, R. (2010). Measuring regional capacity for innovation. Incontext, 11(1). http://www.incontext. indiana.edu/2010/jan-feb/article1.asp Solow, R. M. (1957). Technical change and the aggregate production function. The Review of Economics and Statistics, 39(3), 312-320. http://www.jstor.org/sta- ble/1926047 Statistics Canada. (2019). Canada’ gross domestic prod- uct (GDP) at basic prices by industry data. Stat- Can. https://www150.statcan.gc.ca/t1/tbl1/en/tv.ac- tion?pid=3610043406 Statistics of Japan. (2019). Japan’ gross domestic product classified by economic activities data. E-Stat. https://www.e-stat.go.jp/en/stat-search/ files?page=1&layout=datalist&toukei=00100409&t- https://doi.org/10.1093/ereh/hew020 https://doi.org/10.1093/ereh/hew020 https://doi.org/10.1080/14735970.2019.1652027 https://doi.org/10.1080/14735970.2019.1652027 https://ec.europa.eu/eurostat/statistics-explained/index.php/Glossary:High-tech_classification_of_manufacturing_industries https://ec.europa.eu/eurostat/statistics-explained/index.php/Glossary:High-tech_classification_of_manufacturing_industries https://ec.europa.eu/eurostat/statistics-explained/index.php/Glossary:High-tech_classification_of_manufacturing_industries https://ec.europa.eu/eurostat/statistics-explained/index.php/Glossary:High-tech_classification_of_manufacturing_industries https://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=nama_10_a64&lang=en https://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=nama_10_a64&lang=en https://doi.org/10.1093/icc/dtaa019 https://doi.org/10.1093/icc/dty075 https://doi.org/10.1080/10438599.2018.1557405 https://doi.org/10.1080/10438599.2018.1557405 https://doi.org/10.1787/66a3bd38-en https://doi.org/10.1787/66a3bd38-en http://www.materialflows.net/ http://www.materialflows.net/ https://www.oecd.org/science/inno/39374789.pdf https://www.oecd.org/science/inno/39374789.pdf https://www.oecd.org/sti/ind/48350231.pdf https://www.oecd.org/sti/ind/48350231.pdf https://stats.oecd.org/Index.aspx?DataSetCode=MSTI_PUB https://stats.oecd.org/Index.aspx?DataSetCode=MSTI_PUB https://data.oecd.org/gdp/gross-domestic-product-gdp.htm https://data.oecd.org/gdp/gross-domestic-product-gdp.htm https://data.oecd.org/interest/long-term-interest-rates.htm https://data.oecd.org/interest/long-term-interest-rates.htm https://doi.org/10.1787/9789264304604-en https://doi.org/10.1787/9789264304604-en https://doi.org/10.1080/10438599.2016.1257447 https://doi.org/10.1093/icc/dtz054 https://doi.org/10.1093/icc/dtz054 http://www.incontext.indiana.edu/2010/jan-feb/article1.asp http://www.incontext.indiana.edu/2010/jan-feb/article1.asp http://www.jstor.org/stable/1926047 http://www.jstor.org/stable/1926047 https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=3610043406 https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=3610043406 https://www.e-stat.go.jp/en/stat-search/files?page=1&layout=datalist&toukei=00100409&tstat=000001015836&cycle=7&tclass1=000001090595&tclass2=000001123720&tclass3=000001123721 https://www.e-stat.go.jp/en/stat-search/files?page=1&layout=datalist&toukei=00100409&tstat=000001015836&cycle=7&tclass1=000001090595&tclass2=000001123720&tclass3=000001123721 76 R. Banelienė Journal of Small Business Strategy / Vol. 31, No. 1 (2021) / 66-80 s t a t = 0 0 0 0 0 1 0 1 5 8 3 6 & c y c l e = 7 & t - class1=000001090595&tclass2=000001123720&t- class3=000001123721 Stats NZ. (2019). New Zealand’ gross domestic prod- uct – production measure data. Stats NZ Inforshare. http://archive.stats.govt.nz/infoshare/print.aspx?px- ID=86225c6e-f582-41dc-a3b3-7fdaf1797050 Ulku, H. (2007). R&D, innovation, and growth: Evidence from four manufacturing sectors in OECD countries. Oxford Economic Papers, 59(3), 513–535. https://doi. org/10.1093/oep/gpl022 World Bank. (2019). Population data. The World Bank. https://data.worldbank.org/indicator/SP.POP.TOT- L?view=chart https://www.e-stat.go.jp/en/stat-search/files?page=1&layout=datalist&toukei=00100409&tstat=000001015836&cycle=7&tclass1=000001090595&tclass2=000001123720&tclass3=000001123721 https://www.e-stat.go.jp/en/stat-search/files?page=1&layout=datalist&toukei=00100409&tstat=000001015836&cycle=7&tclass1=000001090595&tclass2=000001123720&tclass3=000001123721 https://www.e-stat.go.jp/en/stat-search/files?page=1&layout=datalist&toukei=00100409&tstat=000001015836&cycle=7&tclass1=000001090595&tclass2=000001123720&tclass3=000001123721 http://archive.stats.govt.nz/infoshare/print.aspx?pxID=86225c6e-f582-41dc-a3b3-7fdaf1797050 http://archive.stats.govt.nz/infoshare/print.aspx?pxID=86225c6e-f582-41dc-a3b3-7fdaf1797050 https://doi.org/10.1093/oep/gpl022 https://doi.org/10.1093/oep/gpl022 https://data.worldbank.org/indicator/SP.POP.TOTL?view=chart https://data.worldbank.org/indicator/SP.POP.TOTL?view=chart 77 R. Banelienė Journal of Small Business Strategy / Vol. 31, No. 1 (2021) / 66-80 Appendix Data used for modeling Country Year GDP per capita ($, current PPPs) Share of industry (% of gross value added) Business-financed GERD (% of GDP) Long term interest rate (%) Australia 2014 47639 6.74 n/d 3.66 2015 47351 6.55 n/d 2.71 2016 50263 6.21 n/d 2.34 2017 51994 6.20 n/d 2.64 Austria 2014 48814 18.56 1.47 1.49 2015 49954 18.72 1.52 0.75 2016 51637 18.74 1.66 0.38 2017 53895 18.65 1.70 0.58 Belgium 2014 44720 14.05 n/d 1.71 2015 45739 14.24 1.44 0.84 2016 47366 14.10 n/d 0.48 2017 49526 14.38 1.72 0.72 Canada 2014 45628 10.45 0.78 2.23 2015 44671 11.08 0.74 1.52 2016 45109 n/d 0.71 1.25 2017 46810 n/d 0.65 1.78 Chile 2014 22688 12.38 0.12 4.74 2015 22593 12.77 0.12 4.48 2016 22788 12.00 0.13 4.41 2017 24181 11.48 0.11 4.24 Czech Republic 2014 32265 26.76 0.71 1.58 2015 33701 26.81 0.67 0.58 2016 35234 26.84 0.66 0.43 2017 38037 26.81 0.70 0.98 Denmark 2014 47905 13.67 n/d 1.33 2015 49071 14.29 1.81 0.69 2016 50685 14.98 n/d 0.32 2017 54337 14.41 1.78 0.48 Estonia 2014 28937 16.15 0.53 n/d 2015 29260 16.01 0.60 n/d 2016 30895 16.02 0.60 n/d 2017 33493 15.65 0.56 n/d Finland 2014 41463 16.90 1.70 1.45 2015 42213 17.12 1.58 0.72 2016 43730 17.03 1.56 0.37 2017 46349 17.70 1.60 0.55 78 R. Banelienė Journal of Small Business Strategy / Vol. 31, No. 1 (2021) / 66-80 France 2014 40144 11.47 1.24 1.67 2015 40841 11.66 1.23 0.84 2016 42067 11.51 1.23 0.47 2017 44125 11.22 n/d 0.81 Germany 2014 47190 22.73 1.89 1.16 2015 47979 22.97 1.91 0.50 2016 49921 23.40 1.90 0.09 2017 52574 23.36 2.01 0.32 Greece 2014 26839 9.51 0.25 6.93 2015 26902 9.80 0.30 9.67 2016 27274 10.53 0.40 8.36 2017 28580 10.81 0.51 5.98 Hungary 2014 25518 23.15 0.65 4.81 2015 26356 24.41 0.68 3.43 2016 26852 23.49 0.68 3.14 2017 28799 23.16 0.71 2.96 Iceland 2014 45713 12.28 0.70 3.20 2015 48857 11.61 0.79 2.66 2016 52340 10.67 0.80 2.78 2017 55330 n/d 0.77 2.22 Ireland 2014 51126 21.75 0.79 2.26 2015 69147 37.16 0.58 1.11 2016 70616 35.41 0.57 0.69 2017 77679 33.93 n/d 0.79 Italy 2014 36071 15.48 0.63 2.89 2015 36836 16.02 0.67 1.71 2016 39045 16.44 0.71 1.49 2017 40981 16.61 n/d 2.11 Israel 2014 34228 14.23 1.46 2.89 2015 35450 14.53 1.41 2.07 2016 37475 13.99 1.61 1.88 2017 38882 13.76 n/d 1.91 Japan 2014 39183 19.88 2.63 0.52 2015 40406 20.93 2.56 0.35 2016 41138 20.79 2.46 -0.07 2017 41985 20.84 2.51 0.05 Korea 2014 33587 30.15 3.23 3.19 2015 35761 29.76 3.14 2.31 2016 37143 29.49 3.19 1.75 2017 38839 30.41 3.47 2.28 79 R. Banelienė Journal of Small Business Strategy / Vol. 31, No. 1 (2021) / 66-80 Latvia 2014 23802 12.35 0.19 2.51 2015 24726 11.97 0.13 0.96 2016 25843 11.87 0.10 0.53 2017 28378 12.23 0.12 0.83 Lithuania 2014 28174 19.19 0.34 2.79 2015 28910 19.27 0.30 1.38 2016 30300 18.77 0.33 0.90 2017 33325 19.12 0.32 0.31 Luxembourg 2014 100934 4.44 n/d 1.34 2015 102817 5.19 0.60 0.36 2016 104702 5.91 n/d -0.18 2017 107525 5.43 n/d 0.50 Mexico 2014 18168 16.87 0.10 6.01 2015 18438 18.29 0.10 5.93 2016 18969 18.20 0.10 6.19 2017 19655 18.25 n/d 7.25 Netherlands 2014 49233 11.50 1.01 1.45 2015 50302 12.01 0.97 0.69 2016 51340 12.11 1.04 0.29 2017 54504 12.33 n/d 0.52 New Zealand 2014 37061 12.09 n/d 4.30 2015 37158 12.36 0.54 3.42 2016 38784 10.95 n/d 2.76 2017 40121 n/d 0.63 2.99 Norway 2014 65896 7.61 n/d 2.52 2015 60357 7.72 0.85 1.57 2016 57728 7.42 0.88 1.33 2017 62012 7.22 0.90 1.64 Poland 2014 25298 18.91 0.37 3.52 2015 26529 19.86 0.39 2.70 2016 27406 20.44 0.51 3.04 2017 29583 19.28 n/d 3.42 Portugal 2014 28747 13.49 0.54 3.75 2015 29685 13.94 0.53 2.42 2016 31042 14.14 0.57 3.17 2017 32554 14.40 0.62 3.05 Slovakia 2014 28928 21.70 0.28 2.07 2015 29700 22.23 0.29 0.89 2016 30896 22.27 0.36 0.54 2017 32376 22.54 0.43 0.92 80 R. Banelienė Journal of Small Business Strategy / Vol. 31, No. 1 (2021) / 66-80 Slovenia 2014 30847 22.87 1.62 3.27 2015 31649 23.08 1.52 1.71 2016 33191 23.32 1.39 1.15 2017 36163 23.72 n/d 0.96 Spain 2014 33728 13.73 0.57 2.72 2015 35054 13.72 0.56 1.74 2016 36743 13.83 0.55 1.39 2017 39087 14.16 0.58 1.56 Sweden 2014 46573 16.50 n/d 1.72 2015 48437 15.46 1.87 0.72 2016 49084 15.11 n/d 0.52 2017 51405 15.36 2.06 0.66 Switzerland 2014 61902 18.99 n/d 0.69 2015 63939 18.54 2.14 -0.07 2016 64324 18.67 n/d -0.36 2017 66396 18.87 2.26 -0.07 Turkey 2014 23983 18.99 0.39 n/d 2015 25728 18.96 0.39 n/d 2016 26330 18.83 0.44 n/d 2017 28190 19.85 0.48 n/d United Kingdom 2014 40878 9.97 0.80 2.57 2015 42055 10.12 0.82 1.90 2016 42943 10.02 0.87 1.31 2017 44909 10.07 n/d 1.24 United States 2014 54952 12.08 1.69 2.54 2015 56718 12.05 1.70 2.14 2016 57822 11.53 1.74 1.84 2017 59879 11.56 1.77 2.33 Source: Prepared by author, based on OECD data (OECD 2019a, 2019b, 2019c, 2019d).