INVESTMENT IN RESEARCH AND DEVELOPMENT AND THEIR IMPACT ON ECONOMIC GROWTH (GDP GROWTH) OF BOSNIA AND HERZEGOVINA

By investing in research and development, Bosnia and Herzegovina can achieve a competitive advantage in the market and play an important role in supporting research and development, because it can facilitate investments through tax policies and measures, cooperation between science, the state and companies, etc. It is important to note that all attempts to support stronger investments in research and development are fully justified, as they lead to economic growth and technological progress to Bosnia and Herzegovina. However, it is important that incentives are created in a way that not only implies investment growth, but also ensures their efficiency. More plastically, it is not only important how much will be invested in R&D, but also how, that is, what exactly will be invested in. Using theoretical models, numerous empirical studies have been conducted that examine the importance of investment in research and development and their impact on economic growth (GDP growth). Such empirical research differs in the selected variables, the type of statistical analysis and the results. The paper uses regression analysis models for selected macroeconomic indicators in the observed period. The aim of the research is to assess the effects of selected macroeconomic factors, especially the impact of investment in research and development on the growth of gross domestic product (dependent variable), and to answer the question whether investment in research and develo-pment (GERD) in Bosnia and Herzegovina is satisfactory?


THE INTRODUCTORY PART
Research and innovation play an essential role in supporting smart and sustainable growth and job creation. While contributing to the creation of new knowledge, research is crucial for the development of new products and technologies, processes and services that enable higher productivity, industrial competitiveness and ultimately prosperity.
In countries with high productivity per capita, research and innovation, and technology development are the main factors of multifactor productivity 1 . The gap between the EU and the US in the intensity of research and development in the business sector 2 is due to the fact that the EU has less opportunity to develop large companies with a strong intensity of research and development based on scientific and technological achievements, especially in ICT. It is a stylized fact that new companies in the EU are growing slower than in the US and that fewer of them are able to break into the ranks of the world's largest companies 3 . 1 Global Competitiveness Report 2016-2017, World Economic Forum, Geneva, 2017, (https:// www.lazarski.pl) 2 The R&D intensity in the business sector in 2015 was 1.25% in the EU, compared to 1.99% in the US. The level of research and development in the business sector is significantly higher in other major economies: "Korea (3.28%), Japan (2.58%) and China (1.59%)" (https://www.ec.europa.eu) 3 Veugelers and Cincera, "Europe missing yollies" ("Europe lacks young leading innovators"), policy summary, Bruegel, Brussels, 2010.
The economic growth and development of Bosnia and Herzegovina is influenced by numerous factors, the most important being the gross domestic product. The choice of these factors is a subjective question of decision makers and analysts (researchers). "The paper uses four independent variables that are included in regression model, and the order of inclusion in the model is as follows: research and development, trade volume, unemployment and inflation. In addition to them, there are other factors important for the functioning of economies: supply and demand for goods and services, unequal distribution of income, political situation, etc. Likewise, natural factors have an equally important impact on economic growth: climate change, diversifica-tion and sustainable development. In order to quantify the relationship between economic growth and other macroeconomic indicators, we chose an extended model of multiple linear regression using the Enter method and regression procedures, as well as residual analysis. The Enter method gradually includes independent variables, and then one by one the variables are removed from the model (according to the previously set criteria)" 4 .
"The goals of the Strategy for the Development of Science in BiH are the construction and development of Bosnia and Herzegovina as a new, modern society, known as the "knowledge society", in which knowledge is the main creative force in personal, economic, social, in cultural and material progress" 5 .

PREVIOUS RESEARCH
Why would investing in R&D be important for a country in general, and for Bosnia and Herzegovina in particular? First, it should be recalled that it has long been shown 6 (Solov, 1956) that the long-term growth rate of economies is determined by technological progress, with technological progress affecting the efficiency of traditional inputs of production, labor and capital (later human capital was included in the models) "Solov's model of economic growth, often known as Solov-Swan's neoclassical model of growth, because this model was independently discovered by Trevor W. Swan and published in The Economic Record in 1956, allows the determinants of economic growth to be separated into increasing inputs (labor and capital) and technical progress. The reason why these models are called "exogenous" growth models is that the savings rate is considered to be exogenous. Subsequent work derives savings behavior from an intertemporal framework to maximize utility. Using his model, Solov (1957) calculated that about four-fifths of the growth in American production per worker could be attributed to technical progress" 7 .
One of the basic endogenous models of growth is Romer's model, which was presented in the paper Endogenous Technological Change in 1990 8 . In this paper, Romer did not reject Solov's explanation of long-term growth, but developed a model that explained how investing in research and development affects technological progress and what influences incentives for entrepreneurs to invest in research and development.
(financing costs play a big role).
www.japmnt.com Since then, Romero's model is one of the most commonly used theoretical models to explain the importance of investing in research and development. Using this theoretical model in a series of empirical studies to date, we conclude that investing in research and development has a positive impact on economic growth. This message is also important for Bosnia and Herzegovina, because the contribution of technological progress to growth is extremely small (average: 0.202% of GDP, period 2005-2019).
In his paper, Mervar 9 states that models of endogenous growth see economic growth as an endogenous result of the economic system, in contrast to the neoclassical model in which economic growth is determined by an exogenously determined rate of technological progress.
Malešević and Ćorić 10 conclude that in the model of endogenous growth related to research, the equilibrium GDP growth rate per worker is determined by investing in research and development. This conclusion means that economic policy makers should encourage investment in research and development. The equilibrium GDP growth rate per worker also depends on the productivity of research. Viewed from a microeconomic perspective, economic policy makers can do little because there are no measures that would, say, affect creating better ideas. Viewed from a macroeconomic perspective, economic policy makers can still to some extent influence the productivity of research through various measures that would affect the quality of education in a country (with emphasis on higher education) and improve the quality of 9 Mervar, A. (1999), A review of economic growth research models and methods. Economic Trends and Economic Policy, 9 (73), p. 20-61. 10  Аttempts to encourage stronger investment in research and development are fully justified, especially if their role is placed in the context of the country's potential growth and technological progress. However, it is important that incentives are created in a way that not only implies investment growth, but also ensures their efficiency.

RESEARCH METHODOLOGY
The research uses a model of single and multiple regression analysis using the databases of the "Statistical Office of Bosnia and Herzegovina" 11 for the observed period (2005 -2019).
Gross domestic product (dependent variable) is analyzed, which is in the function of research and development, volume of trade (market openness), value of high-tech products in total exports, as well inflation and unemployment for the observed period, (T = 15 and one country N = 1). Statistical data are taken from the database of the (http:// www. redete.org) the internet (http://www.bhas.ba/ Calendar/) and Eurostat (vhttps:// ec.europa. eu/eurostat/databrowser/view/rdegerdtot/ default/table?).
Model analysis and comparison is performed using IBM SPSS Statistics v21.

Dependent variable
In the single and multiple linear regression model, the value of GDP was analyzed as a dependent variable. "Gross domestic product. GDP -gross domestic product represents the total value of all final goods and services produced in a given period of time in a given territory. GDP is expressed in monetary units, and includes only goods and services intended for further processing and production.
www.japmnt.com GDP indicates a measure of production activities and is a generally accepted indicator of the state of a particular economy" 12 .

Independent variables
In a simple linear regression model, the value of GDP was tested as a dependent variable and the value of investment in research and development was tested as an independent variable. In the multiple linear regression model, the variables analyzed as independent are:  R&D -Investing in research and development is a proxy variable for technology. The most important indicator of the distribution is gross domestic expenditure on research and development -GERD.
 TRG -Trade in goods means a proxy variable of market openness (exports + imports);  NZP -Unemployment, should answer the question whether it has an impact on GDP growth / decline?
 INF -Inflation in Bosnia and Herzegovina should answer the question of what impact it has on the GDP (growth / decline) ?

RESEARCH RESULTS
Gross domestic product is expressed according to the production approach in millions of EUR ( National Accounts (ESA 95)" 13 , and (SNA 2008) 14 , international standards and recommendations was applied. GDP is the most frequently used aggregate (SNA) and is indicator of overall economic activity. We can conclude that in the observed period, the most significant increase in GDP compared to the previous year was in 2018, when GDP grew by 6,59%. Based on the above, the application of the single linear regression model in this study can be expressed as follows: The following data are used in the application of the single linear regression model ( Table 2): The correlation coefficient of 0.6691 indicates a moderately strong influence (0.5≤ r ≤ 0.8) of the R&D on the GDP. Correlation is positive, and the linear regression shows an yes increase in investment in R&D of EUR 1 million increases GDP by EUR 116.53 million.
The inclusion of data in the statistical software IBM SPSS v21 confirmed the results of the research (Table 3), because the correlation coefficient is (r = 0.668 ≈ 0.6691 and R 2 = 0.447≈0.4477). A positive correlation was also confirmed. The approximate values of the parameters a (a = 10334,808 ≈10332.34) and b (b = 116,434≈116.53) and the linear regression diagram (Table 4 and Graph 2) were also confirmed.   Statistics v21, the Analysis of standard multiple regression was performed, which determined the dependence between the observed variables, based on the database shown in Table 5. It should be emphasized that using this module, multiplex linear regression, will be determined individual relationships and relationships between GDP (dependent variable) on the one hand and independent variables on the other (R&D, TRG, NZP and INF).
Based on the output, attention will be based on the output tables. In the model of standard multiple regression, it is necessary to check the collinearity of the variables, ie. do the independent variables show a weak relationship with the dependent variable (Table 6) www.japmnt.com Table 6: Pearson correlation coefficient values between all variables (Author, 2020).
"It is very important for investors that the country is open to the market, that is that there are no trade restrictions. Therefore, greater market openness is expected to attract more domestic and foreign R&D investment. Research conducted in this paper leads to the conclusion that market openness has a strong impact on the inflow of investment in R&D. R&D represents gross domestic expenditure on research and development (GERD) as a % of GDP, which is a variable for technology" 20 .
Research indicates that greater investment in R&D leads to a significant increase in GDP. Reducing unemployment and inflation will also increase GDP. "Using the Enter method, where the initial situation implies that all independent variables are included in the model, without any restrictions. Then one by one the variables are excluded from the model according to F for removal statistics" 21 . When evaluating the results (small sample) it is necessary to consider the Adjusted R-square, the value of the F-test in testing the significance of all independent variables in the equation, the corresponding p-value of the F-test and the average regression error or standard regression deviation. The more representative the model, the closer the coefficient of determination is expected to be 1, for the higher the value of the F-test, lower the standard deviation of the regression. The following Table 7 shows all the indepe-ndent variables retained in the regression model, a according to predefined statistical criteria. Next Table 8 shows, Model Summary, and key information about the validity of the regression model.   Linear regression shows that an increase in investment in R&D of EUR 1 million increases GDP by 116.53 million eurs.
Research showed that the Square R is 0.965, which means that 96.5% of the variability of the GDP could be explained by the effect of variables: research and development, market openness, unemployment and inflation, while other factors are negligible (3.5%).
It's research confirmed the hypothesis: Investments in R&D have a positive impact on growth GDP, which will encourage structural reforms and enable Bosnia and Herzegovina to adapt to the EU economy. By focusing on new, knowledge-and technology-based products, and increasing the export of hightech products, positive effects on economic growth in Bosnia and Herzegovina will be achieved.
The results of the research can be useful to the Institutes for the Development of Economic Development Strategies and Economic Policies for making optimal decisions in order to resolve any shocks and crises, so that their activities are more precisely directed towards economic growth and to overall progress of BH economy.