171 Mehmet Adak INNOVATIVE DEVELOPMENTS AND ECONOMIC GROWTH: THE CASE OF TURKEY LEARNING OBJECTIVES Once you have mastered the materials in this chapter you will be able to: • Understand the relations between endogenous technological variables and economic growth. • Identify basic technology investments and their returns to aggregate production. • Clarify the key functional relationships between progress in technology and booming economies. • Recognize human capital input in R&D activities. • Analyze the cooperation of human capital and physical capital in the R&D sector. CHAPTER OUTLINE The study of the relationship between technology and economic growth has been the main focus in modern economic literature for over the last century. The Schumpeterian approach was used to explain the economic growth of the developed countries. New technical developments led to cost cuts and price deductions and this innovative competition led to the quick economic growth of countries which encouraged this innovation and development. This chapter analyzes the effects of innovation and technology on economic growth in Turkey over the past three decades. Significant and direct relations have been found between the innovation data and economic growth. We begin with a specific analysis of each series. The series’ graphs are drawn and their descriptive statistics are 172 Mehmet Adak calculated. The relation between innovation and economic growth is tested by Ordinary Least Square Method and Engel Granger test. The test results of the specified period show a positive influence of technological progress on economic growth in Turkey both in the short-run and longrun. KEYWORDS Development, Economic Growth, Innovation, R&D, Technological Progress, Time Series Analysis 1 INTRODUCTION The Turkish economy has always been open to international markets and has always attempted to fully integrate into the global economy both during the Ottoman Empire and Turkish Republic periods. However, the Turkish economy has also been fragile because of political instability and insufficient law systems stemming from poor infrastructure investments. The Turkish economy had a very small growth rate between 1923 and 1939. From World War II up until 1950, the economic growth rate fluctuated frequently. The first multi-party election was in 1950. The liberal party gained power and continued good relations with the allies. Good relations with USA brought new investment opportunities to Turkey. The foreign trade increased immediately. New financial aids and credits were received. The infrastructure investments led to the construction of roads, dams and refineries. Education became more accredited with the west. The army revolted in 1960 and the first election took place in 1962. The 1960s also witnessed the cold war years and the founding of The State Planned Office. The central government focused on planned investments. Its duty was to plan government investments and be a guide for private sector investments. 2 LITERATURE REVIEW Schumpeter was the first modern economist who paid attention to technological progress. He focused on the relation between technological development and economic growth. Today’s R&D growth models are derived from Schumpeter’s theory (Aghion et al. 2014). The backbone of Schumpeterian growth model depends on innovations. New 173 Innovative Developments and Economic Growth:The Case of Turkey innovations are monitored by entrepreneurs who as a result, redirect their investments. Thereby, new technology eliminate old technology (Aghion et al. 2015). The new technology allow producers to produce their yields at lower cost (Romer 1987). A much higher amount can be produced with the same input by using new technology (Romer 1990a). Today new innovations are mainly counted by patent applications. A new piece of literature called, “The Patent Race” describes this competition in detail (Aghion & Howitt 1992). Competition also adds efficiency to the market. Investors aim to invest their funds in fields which have high returns. Consequently, this encourages entrepreneurs to invest in high productive and high profitable sectors (Borsch-supan & Romer 1998). Productivity increases could be observed continuously in the long-run (P. M. Romer 1986a). In addition, technology research also increases consumers’ welfare (Loo & Soete 1999). The results of each research bring about new types of products. Product differentiation allows consumers to have more opportunities. When Technological progress is combined with human capital, the productivity growth can reach much higher levels (Lucas 1988). There is a debate in literature whether technology is exogenous or endogenous (Mankiw et al. 1995). However, most economists lean towards the theory that it is endogenous. Modern technological growth models attempt to prove that technology is a variable which comes from within the model. Economists began studying technology functions in the late 1970s (Romer 2001). These growth functions were thence named new endogenous growth models. The economists aimed to explain countries’ developments by using these new growth models. Technology was accepted as a public possession at the beginning (Romer 1990b). However, the following years saw the monopolization of new technology. New inventions supply monopolistic rent to their owner. These new inventions were protected by patent laws. However not every new invention was kept by patent law. A new term called “intellectual property rights” was now being used in economic literature. Technology and legal systems began to be discussed in the economic system (Romer 1990c). Technology specification found a new platform to be defined again. Economists found that each type of technology had a different kind of influence on economic growth (P. Romer 1986). The following researches focused on the technology change rate. It was observed 174 Mehmet Adak that the countries with a high technology change rate had a higher growth rate than other countries. Thus countries with fast developing technologically had comparative advantages over others (P. M. Romer 1986b). The New technology definition brought about a new aspect in growth theory. The dynamic growth model was built using this new technological definition (Romer 1994). Technology became equivalent to change in today’s world. Globalization and technological progress were now aligned together (Romer 2010). Fig. 1 GDP (Current USD, million USD) Source: Turkish Statistical Institute The Turkish economy became more integrated into the western economic system after 1970. Close relations with USA and Western Europe provided an accelerated force for economic growth. However, despite the economic growth, the law system and infrastructure was not sufficient enough to sustain the growth. In general, the coalition governments were usually in power and this lead to a lack of cooperation between the different political factions which lead to instability. The liberal party won the general election in 1983 and stayed in power for the next 10 years. This period was a breath of fresh air for the economy where telecommunication systems, high ways, integrated electric networks and new hydraulic energy power dams were set up. Furthermore, the new acts were designed for the market. The Turkish currency became convertible and the new capital market act was accepted. New regulations were put forward for foreign investments 175 Innovative Developments and Economic Growth:The Case of Turkey and the new tariff system was checked every week by the cabinet. The Turkish economy grew rapidly in this period and the Foreign trade increased. However, the economy slowed down in the 1990s. The political instability had a very bad effect on the economy. The coalition governments had big budget deficits. Public debts were financed by bonds. The interest rate in the market increased dramatically. The investments rate slowed down. Export and income growth slowed down as well. The exports could not compensate for the imports any more. There were two big devaluations in 1994 and 1999. Fig. 2 GDP per Capita (in Current USD) Source: Turkish Statistical Institute A majority government won the election in 2002. The new government had good relations with the European Union. The law, trade, education, transportation systems were up to standards which are a requirement for the EU. The national technology policy began in 2002 and the technology of national defense was renewed. Private investors were now permitted to invest in defense technology. The National Research and Development law was also accepted in parliament and this led to significant financial support for the R&D investors. In addition, the transportation system policy was renewed with plans for new railways and high ways and constructions (international airports and bridges) was on the way. 176 Mehmet Adak 3 ECONOMIC GROWTH OF THE TURKISH ECONOMY 3.1 National Income Growth The economic growth of modern Turkey can be divided into four periods. The first period is between 1970 and 1983. The 70s were the years when the private sector struggled to join the global economy. The Turkish lira was a pegged currency and the transaction of foreign exchange was not permitted in the domestic market. This led to a strong foreign currency demand in the market. The Turkish lira was losing sudden values. GDP was 18.825 billion USD in 1970. It reached 67.457 billion USD in 1980. GDP had declined to 11 % in 1971 and 17 % in 1980. The economy had grown more than 3 times in this decade in Current USD term. Fig. 3 GDP (Current USD, million USD) Source: Turkish Statistical Institute Table 1 Descriptive Statistics of GDP (million USD) Mean 251351.0 Maximum 823044.0 Median 149195.0 Minimum 16847.00 Std. Dev. 257832.2 Skewness 1.163792 Observations 45 Kurtosis 2.880826 177 Innovative Developments and Economic Growth:The Case of Turkey The army took over the administration in 1980. The next election was in 1983. The economy experienced a 17 % loss in 1980. The growth rate rose 4% in 1981 but the following three years saw a decline in growth rates once again. The growth rates were -10% in 1982, -5% in 1983 and -3% in 1984. The growth rates did not become negative until 1991. The GDP was 67.457 billion USD in 1980 and reached 149.145 billion USD in 1990. The growth of the economy doubled during this decade. Fig. 4 Annual Growth Rate (Current USD % changes) Source: Turkish Statistical Institute The 90s witnessed the coalition years in Turkey. Any political party could not be elected for majority government between 1992 and 2002. Coalition governments usually had public finance problems. Big budget deficits brought high interest rates. Investments slowed down. The Central Bank was known as the finance authority during this period. People used foreign exchange in the domestic market because of high inflation. Three substantial crisis occurred in this decade. One in 1994, one in 1999 and the last one in 2001. The economy shrunk to 26 % in 1994, 9 % in 1999 and 17 % in 2001. All three of the crisis were caused by internal dynamics. There were no global negative influences on the economy. The GDP was 149.156 million USD in 1991. This value could only reach 196.736 USD in 2001. Overall, the economy had grown 30 % in this decade. The last period is 2002 and onwards. There has only been one crisis since which occurred in the year 2009. It was a global crisis. The Turkish economy reduced 17 %. The GDP was 230.494 billion USD in 2002. At 178 Mehmet Adak end of 2014, the GDP was 800 billion USD. The economy had grown three and a half times during this decade. 3.2 GDP per Capita Growth The GDP per capita growth trend has followed the same growth rate of the GDP. Equal fluctuations in each period of GDP can be seen in GDP per capita growth path. Fig. 5 GDP per Capita (Current USD) Source: Turkish Statistical Institute GDP per capita grew 3 times between 1970 and 1980. GDP per capita was 533 USD in 1970. It reached 1518 USD in 1980. Between 1980 and 1990 GDP per capita growth rate was 75 %. GDP per Capita was 2655 USD in 1990. GDP per capital had only increased 16 % between 1990 and 2001. GDP per capita was only 3019 USD in 2001. Per capita income grew more than 3 times in the last decade and it reached 10,404 USD in 2014. 179 Innovative Developments and Economic Growth:The Case of Turkey Fig. 6 GDP per Capita (in USD PPPs) Source: Turkish Statistical Institute 3.3 Manufacturing Sector Growth The manufacturing sector had a slow positive growth rate until 1983. However, from 1983 onwards, the growth rate of the manufacture sector began to increase. The high growth rate slowed down again in the 90s because of coalition governments. This lead to big fluctuations evident in this period. The next break point was the year 2002. The growth rate in the sector had risen more than ever before. But the global crises effected the sector in 2009 negatively. The sector is still struggling to recover. There has been a positive growth rate since 2010 but not as high as before. Fig. 7 Manufacturing, value added (current US$) Source: World Bank, World Development Indicators Data Base 180 Mehmet Adak The new liberal government encouraged exporters by the new value added tax return law in 1983. The foreign trade tariffs were designed to attract exporters. Raw material and intermediate input imports became easier. Foreign exchange currencies were now permitted to be transacted in the market. The price of currencies was determined in support of the exporters. Fig. 8 Manufactures Export and Imports Source: World Bank, World Development Indicators Data Base The share of manufacture exports in merchandise exports has increased in the past decades. This increase is perceived as a diminished increase. 26.8 % of exports was sourced from the manufacture sector. The share increased to 67.8 % in 1990 and 84.5 % in 2004. The share increase stopped in 2004 and shares have begun to decrease slightly since with absolute values. The share reduced to 77.73% in 2013. The structure of manufacture import share in merchandise imports is similar to that of manufacture exports, however the increase halted in 1998 with 75.6 %. The rate reduced to 57.8 in 2013. The trend shows that the structure of Turkish imports changed dramatically after 1998. Turkey is not importing manufactured goods as it previously did. 4 INNOVATIVE DEVELOPMENT Most of the Turkish Technological statistics have been recorded since 1990. As such, there is a lack of secure data sources before 1990. This section is planned to shed some light on these limited resources. 181 Innovative Developments and Economic Growth:The Case of Turkey 4.1 Gross Domestic Expenditure on R&D The ratio of each country’s Gross Domestic Expenditure (GERD) on R&D to GDP is usually accepted as one of the prime indicators. Turkey’s GERD as a percentage of GDP was 0.24 % in 1990. This ratio reached 0.48 % in the year 2000. GERD as a percentage of GDP was recorded as 0.95 %. Figure 9 gives information of world developed countries’ R&D expenditures. Turkey’s expenditure ratio is less than other countries on the graph. But there is an increasing trend for Turkey. It has almost doubled over the last ten years. The GERD of Germany as a percentage of GDP was 2.94 %, France was 2.23 %, and UK was 1.63 % in 2013. Fig. 9 GERD as Percentage of GDP Source: OECD Main Science and Technology Indicators Turkey’s GERD was about 780 million USD in 1990. It increased four times in the following 10 years and reached 2.8 billion USD in the year 2000. At the end of 2013, GERD was 13.3 billion USD in Turkey. The world highest R&D expenditure was recorded in USA in 2013. It was about half a trillion USD. Germany’s R&D expenditure was over 100 billion USD, France’s was 55 billion USD and UK’s was 40 billion USD in 2013. 182 Mehmet Adak Fig. 10 GERD (million USD) Source: OECD Main Science and Technology Indicators Financial source of R&D expenditure has changed. The percentage of central or federal governments in each country’s R&D expenditures have declined over the last three decades. The government’s share was nearly 50 % in developed countries. It was 53.4% in France, 47.8% in USA, 46.37% in UK and 41.79% in Germany at the beginning of the 80s. Central governments’ financing shares decreased radically in the following years. The developed countries central government R&D expenditures reduced to around 40 % in 90s and 30 % after the year 2000. The ratios of central government finance on R&D in France was 34.97%, 30.97% in US, 26.99% in UK and 29.21 % in Germany in 2013 Turkey’s public share of R&D expenditure was 71.42% in 1990. It decreased to 50.6 % in 2000 and was recorded as 26.55 % in 2013. 183 Innovative Developments and Economic Growth:The Case of Turkey Fig. 11 Percentage of GERD Financed by Government Source: OECD Main Science and Technology Indicators Percentage of R&D financed by private industry have increased over the whole world, both in developed and developing countries. Industry contribution on R&D expenditure in many countries was below the %50 in 1980. It was 40.92 % in France, 42.05 % in UK and 49.42 % in US. Germany was a bit higher with a ratio of 56.85%. Fig. 12 Percentage of GERD Financed by Industry Source: OECD Main Science and Technology Indicators By the time Germany’s Industry contribution on R&D expenditure reached to 66.07 % in 2013, the ratio was announced as 59.13 % in 184 Mehmet Adak US, 55.38% in France and 46.55 % in UK. More than half of R&D expenditure is financed by industry in most of the developed countries. Turkey’s industry ratio in R&D expenditure was 27.44 % in 1990. It increased to 42.92 % in 2000 and 48.87 % in 2013. Fig. 13 Total Researchers per Thousand Total Employment Source: OECD Main Science and Technology Indicators The ratio of total Turkish researchers to total Turkish employment was 0.067 percent in 1990 while it was 0.6 % in USA and in Germany, 0.5 % in France and the UK. The structure of trend changed for Turkey in 2002. It became steeper while other countries remained the same. The ratio of researchers to employment was 0.34 percent for Turkey in 2013. The developed countries’ ratios were around 0.8 – 0.9 percent in 2013. 89 thousand Turkish researchers were employed in the market at the end of 2013. This statistic recorded 1.3 million in USA, 360 thousand in Germany, 265 thousand in France and 259 thousand in UK. 185 Innovative Developments and Economic Growth:The Case of Turkey Fig. 14 Scientific and Technical Journal Articles Source: World Bank, World Development Indicators Data Base The total number of universities was 184 in 2013. It was 28 in 1985 and 73 in 2000. The increase in the number of universities had a positive effect on new job opportunities for researchers in universities. The new and increasing positions in universities had a direct influence on published scientific papers. The number of total published scientific paper was 326 in 1985 and was doubled in 1990 to 750. This increased to 3,484 in 2000 and 8,328 in 2013. 208.6 thousand Scientific papers were published from USA in 2011. Germany and UK had an annual of 46 thousand scientific papers and France’s papers were recorded as 31.6 thousand. 186 Mehmet Adak Fig. 15 Patent Applications in Turkey Source: World Bank, World Development Indicators Data Base The total resident patent applications were 134 in 1980. It had a horizontal trend until 2000 when the total number of annual application reached 277. After 2000 the annual application growth rate increased. 2,268 applications were completed in 2008 and 4,861 applications in 2014. Fig. 16 Trademark Applications Source: World Bank, World Development Indicators Data Base Non-resident applications increased until 1995. Total non-resident applications were 1,520 in 1995. However, this reduced to 713 in 1996 and rose again to 3,156 in 2000. This increase was followed by a decline until 2003. The total non-resident applications were 685 in 2003. The 187 Innovative Developments and Economic Growth:The Case of Turkey following years were increment years. Total non-resident patent applications were 7,514 in 2014. The total number of Turkish firms’ trademark applications to Turkish Patent Office were 2,143 in 1980. 1980s were very poor years for direct resident applications. The total number of Turkish brands or trademarks applications began to increase in the 90s. The total nonresident application was 3,200 in 1990. The year 2000 and the following years marked a speed up for registration. Total non-resident applications reached 8,344 in 2000 and14,746 in 2013. The crisis of 1994, 1999, 20001 and 2009 could be easily noticed in resident applications time path. 93,341 trademark applications were completed by resident mark owners to National Patent Office in 2013. 4.2 Technology Transfer Turkey’s computer and electronic exports was 4.6 million US Dollar in 1981. The new regulations which aimed at integrating the national economy with the global economy gave its fruit in the following ten years. Turkish computer and electronic exports rose to 302.6 million US Dollars in 1990. Technology imports increased more than three times in 2000 and it was recorded 1.074 billion US Dollar. This figure rose to 2.66 billion US Dollars in 2013. Fig. 17 Export of Computer, Electronic and Optical Industry (Current USD in million) Source: OECD Main Science and Technology Indicators 188 Mehmet Adak Turkey entered the arena of international technology competition late. Turkish technology exports statistics in US Dollar is quite low but exports growth rate is comparatively higher than developed countries. In 2013, technology exports for USA was 194.6 billion US Dollars, Germany was 114 billion US Dollars, France was 37.1 billion US Dollars and UK was 35.8 billion Dollars. Fig. 18 Import of Computer, Electronic and Optical Industry (Current USD in million) Source: OECD Main Science and Technology Indicators Computer and electronic devices imports provided technology transfers. Turkish computer and electronic imports was 177.7 million US Dollars in 1980. Technology transfer increased with the new liberalization in the economy. Computer and electronic imports increased ten time reaching 1.66 billion US Dollars in 1990. It was 6.4 billion US Dollars in 2000 and 14.5 US Dollars in 2013. Turkish technology imports were almost quarter of France, UK, half of Germany and one tenth of USA in 2013. There has been another significant development in Turkey which is that of the aerospace industry. Turkey’s aerospace export was zero in the 1980s. Turkish aerospace exports was 0.54 million in 1990. Turkish Aerospace Industry (TAI) was founded in Ankara. Turkey started to produce fighter jets in Ankara with NATO allies. USA is still a strategic partner in the aerospace industry. The industry developed rapidly and industry exports increased to 582.1 million US Dollars in 2000. In 2013 the aerospace industry export was 931.7 million US Dollars. The Turkish 189 Innovative Developments and Economic Growth:The Case of Turkey aerospace industry exports is however behind that of the exports of developed countries. Fig. 19 Total Exports of Aerospace Industry (Current USD in million) Source: OECD Main Science and Technology Indicators US aerospace exports was 116.8 billion US Dollars, France was 67.1 billion US Dollar, Germany was 51 billion US Dollars and the UK was 37.7 billion US Dollars in 2013. Fig. 20 Total Imports of Aerospace Industry (Current USD in million) Source: OECD Main Science and Technology Indicators Turkish aerospace imports is comparatively behind developed countries. Turkish aerospace imports was 23 million US Dollars in 1980, it rose to 190 Mehmet Adak 298.3 million US Dollars in 1990. The year 2000 statistics is 1.1 billion US Dollars and it reached 2.6 billion US Dollars in 2013. UK imports was 30 billion US Dollars and USA was 46 billion US Dollars in 2013, while Germany’s import was 35.5 billion US Dollars and France’s imports was 38 billion US Dollars. 5 THE ANALYSIS AND RESULTS In this section structural determinants of technological development in Turkey is analyzed. Each variable given in the previous section is classified into three classes. The linear relations between these classes is then tested by ordinary least square methods. 5.1 Variables The data set is classified into two classes as Economic Growth variables and Innovation sector variables. Table 2 Classification of the Variables Classes Variables Economic Growth Gross Domestic Product Gross Domestic Product per Capita Innovation Sector Gross Domestic Expenditure on R&D Total Researchers Scientific and Technical Journal Articles Patent Applications Trade Mark Applications Export of Computer, Electronic and Optical Industry Export of Aerospace Industry Import of Computer, Electronic and Optical Industry Import of Aerospace Industry 191 Innovative Developments and Economic Growth:The Case of Turkey The series were supplied from different sources. The Gross Domestic Product and GDP per Capita series were sourced from the Turkish Statistic Institute. The series are published in current US Dollars. Gross Domestic Expenditure on R&D and Total Researchers statistics were sourced from OECD Main Science and Technology Indicator Data Base. Gross Domestic Expenditure on R&D is published in current US Dollar. Total Number of Scientific and Technical Journal Articles, Total Patent Applications, Total Trade Mark Applications were taken from World Bank Development Indicator Data Base. Total Export - Import of Computer, Electronic and Optical Industry and Total Export-Import of Aerospace Industry series were sourced from OECD Main Science and Technology Indicator Data Base. Both series are in term of current US Dollar. All series are run in natural logarithm form for the analysis. The natural logarithm series of all variables, their descriptive statistics and their distributions are given below; 192 Mehmet Adak Fig. 21 Time Series of All Variables in Natural Logarithm 193 Innovative Developments and Economic Growth:The Case of Turkey Fig. 22 Distribution of All Variables in Natural Logarithm 194 Mehmet Adak Ta bl e 3 D es cr ip tiv e St at is tic s of A ll Va ria bl es ln G DP ln G DP p er Ca pi ta ln G ro ss D om es tic Ex pe nd itu re on R &D ln To ta l N um be r o f R es ea rc he rs ln Sc ie nt ifi c a nd Te ch ni ca l J ou rn al A rt icl es M ea n 26 .0 55 37 8. 93 78 47 21 .8 98 85 10 .4 64 64 7. 77 61 03 M ed ia n 25 .9 42 90 8. 97 68 81 21 .7 96 84 10 .2 51 46 7. 93 55 87 M ax im um 27 .4 36 28 9. 82 95 18 23 .3 12 16 11 .6 34 87 9. 06 42 39 M in im um 24 .7 94 73 7. 96 41 70 20 .4 73 34 9. 54 33 06 5. 78 68 97 S td . D ev . 0. 87 40 79 0. 53 05 54 0. 86 48 88 0. 67 63 66 1. 11 45 51 S ke w ne ss 0. 16 32 10 0. 01 23 17 0. 18 68 52 0. 30 22 94 -0 .3 36 19 2 K ur to si s 1. 78 74 67 2. 06 28 56 1. 80 59 08 1. 73 69 84 1. 72 59 78 Ja rq ue -B er a 2. 23 37 80 1. 24 50 31 1. 56 55 10 1. 96 07 37 2. 33 46 37 P ro ba bi lit y 0. 32 72 96 0. 53 65 93 0. 45 71 45 0. 37 51 73 0. 31 12 00 S um 88 5. 88 24 30 3. 88 68 52 5. 57 25 25 1. 15 15 20 9. 95 48 S um S q. D ev . 25 .2 12 47 9. 28 90 94 17 .2 04 70 10 .5 21 85 32 .2 97 80 O bs er va tio ns 34 34 24 24 27 195 Innovative Developments and Economic Growth:The Case of Turkey ln Pa te nt A pp lica tio ns ln Tr ad e M ar k Ap pl ica tio ns ln Ex po rt of Co m pu te r, El ec tr on ic an d Op tic al In du st ry ln Im po rt of Co m pu te r, El ec tr on ic an d Op tic al In du st ry ln Ex po rt of A er os pa ce In du st ry ln Im po rt of A er os pa ce In du st ry M ea n 5 .8 95 06 4 9 .5 53 38 2 1 9. 81 30 5 2 1. 73 61 1 1 7. 55 99 4 1 9. 94 46 0 M ed ia n 5 .2 77 47 6 9 .5 90 96 6 2 0. 24 23 1 2 1. 95 62 1 1 9. 15 55 0 2 0. 37 55 1 M ax im um 8 .3 97 05 7 1 1. 54 97 2 2 1. 92 83 0 2 3. 40 03 4 2 0. 65 26 1 2 2. 14 98 4 M in im um 4 .8 36 28 2 7 .5 93 87 8 1 5. 23 37 9 1 8. 99 59 4 1 0. 59 66 3 1 5. 00 94 3 S td . D ev . 1 .1 58 51 7 1 .1 26 48 2 2 .0 41 56 9 1 .2 14 47 1 3 .4 13 96 1 1 .7 08 65 6 S ke w ne ss 1 .0 10 89 6 -0 .0 24 40 1 -0 .9 39 59 5 -0 .5 47 10 3 -0 .8 80 92 7 -1 .3 91 42 0 K ur to si s 2 .5 31 81 6 1 .9 25 10 4 2 .8 97 37 2 2 .4 15 39 2 2 .2 08 33 8 4 .8 02 32 1 Ja rq ue -B er a 5 .7 42 45 1 1 .4 95 47 0 4 .8 70 09 3 2 .1 16 20 0 4 .5 08 12 0 1 5. 11 47 6 P ro ba bi lit y 0 .0 56 62 9 0 .4 73 43 8 0 .0 87 59 4 0 .3 47 11 5 0 .1 04 97 2 0 .0 00 52 2 S um 1 88 .6 42 1 2 96 .1 54 8 6 53 .8 30 5 7 17 .2 91 7 5 09 .2 38 3 6 58 .1 72 0 S um S q. D ev . 4 1. 60 70 3 3 8. 06 88 8 1 33 .3 76 1 4 7. 19 80 9 3 26 .3 43 6 9 3. 42 41 3 O bs er va tio ns 3 2 3 1 3 3 3 3 2 9 3 3 196 Mehmet Adak 5.2 The Unit Root Test The stationary series usually fluctuate around a fixed mean with a fixed variance. Augmented Dickey-Fuller Test is run for checking stationarity. The test is based on the following regression model; If the series is found non-stationary, then the same test is run with the first difference forms of the same series as; The Dickey-Fuller stationary test is employed for every series. The series were found non-stationary in level and found stationary in first difference. All series’ unit root test outputs in level and in first difference are given below; Table 4 Unit Root Test Results for Series in Level Variable Test for Unit Root in Include in test equation Lag length Augmented Dickey-Fuller Test Statistic Probability (MacKinnon 1996, one sided) Result ln (GDP) level Trend, Intercept Schwarz info Criteria -3.1042 0.1219 It has a unit root ln (GDP per Capita) level Trend, Intercept Schwarz info Criteria -2.5208 0.3167 It has a unit root ln (Gross Domestic Expenditure on R&D) level Trend, Intercept Schwarz info Criteria -2.3676 0.3849 It has a unit root ln (Total Number of Researchers) level Trend, Intercept Schwarz info Criteria -1.9046 0.6196 It has a unit root 197 Innovative Developments and Economic Growth:The Case of Turkey ln (Scientific and Technical Journal Articles) level Trend, Intercept Schwarz info Criteria 1.5491 1.0000 It has a unit root ln (Patent Applications) level Trend, Intercept Schwarz info Criteria -0.8349 0.9507 It has a unit root ln (Trade Mark Applications) level Trend, Intercept Schwarz info Criteria -.32096 0.1023 It has a unit root ln (Export of Computer, Electronic and Optical Industry) level Trend, Intercept Schwarz info Criteria -1.3785 0.8481 It has a unit root ln (Export of Aerospace Industry) level Trend, Intercept Schwarz info Criteria -1.3980 0.8393 It has a unit root ln (Import of Computer, Electronic and Optical Industry) level Trend, Intercept Schwarz info Criteria -3.1367 0.1153 It has a unit root ln (Import of Aerospace Industry) level Trend, Intercept Schwarz info Criteria -2.6263 0.2721 It has a unit root 198 Mehmet Adak Table 5 Unit Root Test Results for Series in First Difference Variable Test for Unit Root in Include in test equation Lag length Augmented Dickey-Fuller Test Statistic Probability (MacKinnon 1996, one sided) Result ln (GDP) First difference Trend, Intercept Schwarz info Criteria -6.0073 0.0001 It does not have a unit root ln (GDP per Capita) First difference Trend, Intercept Schwarz info Criteria -5.9604 0.0001 It does not have a unit root ln (Gross Domestic Expenditure on R&D) First difference Trend, Intercept Schwarz info Criteria -5.3798 0.0014 It does not have a unit root ln (Total Number of Researchers) First difference Trend, Intercept Schwarz info Criteria -5.4601 0.0012 It does not have a unit root ln (Scientific and Technical Journal Articles) First difference Trend, Intercept Schwarz info Criteria -3.6232 0.0481 It does not have a unit root ln (Patent Applications) First difference Trend, Intercept Schwarz info Criteria -7.5711 0.0000 It does not have a unit root ln (Trade Mark Applications) First difference Trend, Intercept Schwarz info Criteria -4.8625 0.0030 It does not have a unit root 199 Innovative Developments and Economic Growth:The Case of Turkey ln (Export of Computer, Electronic and Optical Industry) First difference Trend, Intercept Schwarz info Criteria -3.5212 0.0557 It does not have a unit root ln (Export of Aerospace Industry) First difference Trend, Intercept Schwarz info Criteria -7.7632 0.0000 It does not have a unit root ln (Import of Computer, Electronic and Optical Industry) First difference Trend, Intercept Schwarz info Criteria -5.7907 0.0002 It does not have a unit root ln (Import of Aerospace Industry) First difference Trend, Intercept Schwarz info Criteria -7.2655 0.0000 It does not have a unit root 200 Mehmet Adak 5.3 Ordinary Least Square Models Seven regression analysis were conducted. Each models dependent and independent variable(s) are summarized in the table below: Table 6 Dependent and Independent Variables of Each Model Model Dependent Variable Independent Variable(s) 1st Patent Applications Gross Domestic Expenditure on R&D Total Number of Researchers 2nd GDP per Capita Export of Computer, Electronic and Optical Industry Export of Aerospace Industry 3rd Trade Mark Applications Patent Applications 4th Export of Computer, Electronic and Optical Industry Patent Applications 5th Export of Aerospace Industry Patent Applications 6th Patent Applications Scientific and Technical Journal Articles 7th GDP Import of Computer, Electronic and Optical Industry Import of Aerospace Industry The regression analysis outputs are given below: 201 Innovative Developments and Economic Growth:The Case of Turkey Table 7 First Model’s Output Dependent Variable: ln(Patent Applications) Method: Least Squares Sample (adjusted): 1991 2012 Included observations: 22 after adjustments Variable Coefficient Std. Error t-Statistic Prob. Intercept -16.15260 2.243575 -7.199495 0.0000 ln(Gross Domestic Expenditure on R&D)t-1 0.337428 0.237931 1.418180 0.1723 ln(Total Number of Researchers) 1.445952 0.298371 4.846149 0.0001 R-squared 0.977191 Mean dependent var 6.309403 Adjusted R-squared 0.974790 S. D. dependent var 1.183703 S. E. of regression 0.187943 Akaike info criterion -0.379232 Sum squared resid 0.671129 Schwarz criterion -0.230453 Log likelihood 7.171551 Hannan-Quinn criter. -0.344184 F-statistic 407.0069 Durbin-Watson stat 0.667817 Prob(F-statistic) 0.000000 202 Mehmet Adak Table 8 Second Model’s Output Dependent Variable: ln(GDP per Capita) Method: Least Squares Sample (adjusted): 1984 2012 Included observations: 29 after adjustments Variable Coefficient Std. Error t-Statistic Prob. Intercept 4.028771 0.667591 6.034790 0.0000 ln(Export of Computer, Electronic and Optical Industry) 0.231696 0.048610 4.766379 0.0001 ln(Export of Aerospace Industry) 0.017698 0.021606 0.819159 0.4201 R-squared 0.850647 Mean dependent var 9.022833 Adjusted R-squared 0.839159 S. D. dependent var 0.440599 S. E. of regression 0.176702 Akaike info criterion -0.531003 Sum squared resid 0.811817 Schwarz criterion -0.389558 Log likelihood 10.69954 Hannan-Quinn criter. -0.486704 F-statistic 74.04233 Durbin-Watson stat 0.236355 Prob(F-statistic) 0.000000 203 Innovative Developments and Economic Growth:The Case of Turkey Table 9 Third Model’s Output Dependent Variable: ln(Trade Mark Applications) Method: Least Squares Sample (adjusted): 1980 2011 Included observations: 31 after adjustments Variable Coefficient Std. Error t-Statistic Prob. Intercept 4.215158 0.538032 7.834393 0.0000 ln(Patent Applications) 0.918111 0.091022 10.08674 0.0000 R-squared 0.778190 Mean dependent var 9.553382 Adjusted R-squared 0.770541 S. D. dependent var 1.126482 S. E. of regression 0.539606 Akaike info criterion 1.666386 Sum squared resid 8.444069 Schwarz criterion 1.758902 Log likelihood -23.82899 Hannan-Quinn criter. 1.696544 F-statistic 101.7424 Durbin-Watson stat 0.189767 Prob(F-statistic) 0.000000 204 Mehmet Adak Table 10 Fourth Model’s Output Dependent Variable: ln(Export of Computer, Electronic and Optical Industry) Method: Least Squares Sample (adjusted): 1981 2012 Included observations: 31 after adjustments Variable Coefficient Std. Error t-Statistic Prob. Intercept 12.86412 1.277399 10.07056 0.0000 ln(Patent Applications) 1.186284 0.211607 5.606065 0.0000 R-squared 0.520090 Mean dependent var 19.89550 Adjusted R-squared 0.503541 S. D. dependent var 1.913143 S. E. of regression 1.347998 Akaike info criterion 3.497459 Sum squared resid 52.69585 Schwarz criterion 3.589974 Log likelihood -52.21061 Hannan-Quinn criter. 3.527616 F-statistic 31.42796 Durbin-Watson stat 0.119892 Prob(F-statistic) 0.000005 205 Innovative Developments and Economic Growth:The Case of Turkey Table 11 Fifth Model’s Output Dependent Variable: ln (Export of Aerospace Industry) Method: Least Squares Sample (adjusted): 1984 2012 Included observations: 29 after adjustments Variable Coefficient Std. Error t-Statistic Prob. Intercept 6.166454 2.591204 2.379764 0.0246 ln(Patent Applications) 1.900535 0.424467 4.477461 0.0001 R-squared 0.426114 Mean dependent var 17.55994 Adjusted R-squared 0.404859 S. D. dependent var 3.413961 S. E. of regression 2.633714 Akaike info criterion 4.841139 Sum squared resid 187.2841 Schwarz criterion 4.935435 Log likelihood -68.19651 Hannan-Quinn criter. 4.870671 F-statistic 20.04766 Durbin-Watson stat 0.136177 Prob(F-statistic) 0.000124 206 Mehmet Adak Table 12 Sixth Model’s Output Dependent Variable: ln(Patent Applications) Method: Least Squares Sample (adjusted): 1985 2011 Included observations: 27 after adjustments Variable Coefficient Std. Error t-Statistic Prob. Intercept -0.523632 0.847059 -0.618177 0.5421 ln(Scientific and Technical Journal Articles) 0.831428 0.107869 7.707738 0.0000 R-squared 0.703824 Mean dependent var 5.941637 Adjusted R-squared 0.691977 S. D. dependent var 1.104568 S. E. of regression 0.613033 Akaike info criterion 1.930392 Sum squared resid 9.395248 Schwarz criterion 2.026380 Log likelihood -24.06030 Hannan-Quinn criter. 1.958935 F-statistic 59.40922 Durbin-Watson stat 0.095695 Prob(F-statistic) 0.000000 207 Innovative Developments and Economic Growth:The Case of Turkey Table 13 Seventh Model’s Output Dependent Variable: ln(GDP) Method: Least Squares Sample (adjusted): 1981 2013 Included observations: 33 after adjustments Variable Coefficient Std. Error t-Statistic Prob. Intercept 11.36649 0.886218 12.82584 0.0000 ln(Import of Computer, Electronic and Optical Industry) 0.671984 0.067520 9.952297 0.0000 ln(Import of Aerospace Industry) 0.005842 0.047992 0.121729 0.9039 R-squared 0.908933 Mean dependent var 26.08932 Adjusted R-squared 0.902862 S. D. dependent var 0.864556 S. E. of regression 0.269456 Akaike info criterion 0.301682 Sum squared resid 2.178191 Schwarz criterion 0.437729 Log likelihood -1.977760 Hannan-Quinn criter. 0.347458 F-statistic 149.7145 Durbin-Watson stat 0.302698 Prob(F-statistic) 0.000000 All independent variables’ parameter signs were found positive as they were expected. R2 of all seven analyses are over 50 percent. Export of Aerospace Industry and Import of Aerospace Industry variables’ parameter’ t-statistics values are a low in model 2nd and model 8th. Gross Domestic Expenditure on R&D variable parameter’s t-statistics 208 Mehmet Adak value is significant at the 80 percent confidence level. The Rest of the independent variables’ parameter t-statistics are significant at the 95% confidence level. 5.4 Co-integration Tests The Engel-Granger method is used to check co-integration between variables’ series. This test is usually employed to check long-run relationships between dependent and independent variables. All series were found non-stationary in level and stationary in first difference. If the residual of each OLS model is stationary at level I (0), then a long run relation between variables can be assumed. In other words, the series are co-integrated. The residual time series’ Augmented Dickey Fuller test results are given below: Table 14 All Models’ Residual Unit Root Test Model Test for Unit Root in Include in test equation Lag length Augmented Dickey-Fuller Test Statistic Probability (MacKinnon 1996, one sided) Result 1st Model Level None Schwarz info Criteria -2.4226 0.0180 It does not have a unit root 2nd Model Level None Schwarz info Criteria -0.6313 0.4344 It has a unit root 3rd Model Level None Schwarz info Criteria -2.0936 0.0368 It does not have a unit root 4th Model Level None Schwarz info Criteria -2.0237 0.0429 It does not have a unit root 5th Model Level None Schwarz info Criteria -2.5086 0.0144 It does not have a unit root 209 Innovative Developments and Economic Growth:The Case of Turkey 6th Model Level None Schwarz info Criteria -1.9468 0.0511 It does not have a unit root 7th Model Level None Schwarz info Criteria -3.1597 0.0025 It does not have a unit root 2nd Model’s residual was found non-stationary in level. This means that the regression model results do not work in the long-run. But the rest of the 6 models (1st, 3rd, 4th, 5th, 6th and 7th) residuals were found stationary in level. The variables of each models fluctuate together and have long run relationships. 6 ANALYSIS AND RESULTS Two types of analysis were used during this study. These analyses’ outputs are summarized under short-run and long-run headings. 6.1 Short-Run Analyses There were 7 ordinary least square models used for short-run analysis. Each models’ dependent and independent variables are given in Table 6. The first model’s dependent variable was patent applications in Turkey. The independent variables are gross domestic expenditure on R&D and total number of researchers. Least square model’s output is given in Table 7. The effects of each independent variable on the dependent variable was found positive in the model. The number of researchers’ influence on patent application is almost 4.5 times greater than R&D expenditure. R-square value was found quite high. The second model aims to test the export of computers, electronic, optical industries and the export of aerospace industry on GDP per capita. Both independent variables effects on dependent variables were found positive. Export of aerospace industry could explain why the GDP per capita growth is very limited and the same variable’s t-statistics value is very low. R-square was calculated as 85 percent. The third model was designed to shed light on the interrelation between total trade mark applications and total patent applications. According to ordinary least square model, there is a positive relation 210 Mehmet Adak between two variables. The independent variable’s (total patent applications) t-statistic value was found significant at 99 percent. R-square value was 77 percent. The forth model’s independent variable is total patent applications. The dependent variable is the total export of computer, electronic and optical industry. Total patent application can explain growth in the export of computer, electronic and optical industry. Standard error of independent variable is quite low and the adjusted R-square value is almost 50 percent. The fifth model shows the relation between total number of patent applications and total aerospace industry export in Turkey. The total number of patent application is independent variable in the least square model. Signs of independent variables is found positive as was expected. The number of patent application’s t-statistic value is inside the 90 percent significance boundary. The r-square value is 42 percent. The sixth model aims to check the interrelation between the total number of articles which were published in scientific journals and the total number of patent applications. The total number of published papers is the independent variable of the model. Independent variable’s coefficient was found 0.83 and its standard error is low. The Independent variable can explain the 70 percent changes in the dependent variable. The seventh and last model was used to explain GDP growth by two inputs which are the total import of computer, electronic, optical industry and total import of aerospace industry. Both independent variables’ coefficients signs were found positive. According to coefficients’ magnitudes: the influence of total import of aerospace industry on GDP is less than the total import of computer, electronic, optical industry. The t-statistic value of total import of aerospace is very low while the total import of computer, electronic and optical variable’s t-statistics is in the 90 % confidence interval. The R-square is calculated as 90 percent. 6.2 Long-Run Analyses According to the Engel-Granger method, there can be a long-run relationship between dependent and independent variables if the series of least square model are non-stationary in level, and the least square model’s residual is stationary in level. The residual series of the 7 models 211 Innovative Developments and Economic Growth:The Case of Turkey are tested in long-run analysis. The unit root test is run and the tests results are given in Table 14. Long-run interrelation is found in 6 models out of 7 models. These models model numbers are 1, 3, 4, 5, 6 and 7. These least square model’s residuals were found stationary in level. In other words, there are no unit roots found in those residual series. No co-integration was found in model 2. The residual series of the second least square model was found non-stationary in level which shows that the series has unit root. 7 CONCLUSION The determinants of technological progress and economic growth in Turkey were examined in this paper. It can be seen that all independent variables in the seven models provided, have a positive influence on economic growth in the short run. However, the long-run analysis results suggest that independent variables such as the export of computers, electronics, optical industry and Aerospace industry products do not have a positive effect on GDP per capita. The results indicate that compared to the other models, the first model of Gross Domestic Expenditure on R&D and Total number of researchers has a considerable positive effect on Patent Applications. New research can thus focus on an index which explains the multiple linear relations in the progress of technology. A strong relation was found between total patent applications and total trade mark applications during the years between 1980 and 2011 in model 3. The patent and trade mark applications dramatically increased simultaneously during this period in Turkey. New inventions are patented first in national patent office and the new trade mark applications are done for the new invented product in the same office. A Positive influence of total patent applications on total export of computer, electronic, optical industry was found in model 4. Any increase on patent application had a remarkable effect on total Turkish technologic export. Growing export values were clarified by patent application over the past decade. In addition, patent applications had a positive affect on aerospace industry exports. Aerospace industry has been developed for the last two decades. Furthermore, new technological 212 Mehmet Adak inventions which are registered by national patent office triggered the aerospace industry production and aerospace industry export. Patent application are encouraged by Scientific and technical journal articles. This econometric relation was found to be very significant. Scientific papers which written in Turkey have increased dramatically over the last decades. This increase on the publishing of scientific papers, supported the increase on patent application. Technology import such as computer and aerospace imports has a weak positive effect on GDP growth. Computer and aerospace imports are called basic inputs of technology production. The relation between export of computer-aerospace industries and economic (GDP) growth is econometrically significant in short-run. Nonetheless, no long-run relation was found between these variables. REFERENCES Aghion, P. , Akcigit, U. & Howitt, P. , 2015. Lessons from Schumpeterian Growth Theory. American Economic Review: Papers & Proceedings, 105(5), pp. 94 – 99. Available at: http://dx.doi.org/10.1257/aer.p. 20151067. Aghion, P. , Akcigit, U. & Howitt, P. , 2014. What Do We Learn From Schumpeterian Growth Theory? Handbook of Economic Growth, 2, pp. 515 – 563. Aghion, P. & Howitt, P. , 1992. A Model of Growth Through Creative Destruction. Econometrica, 60(2), pp. 323 – 351. Available at: http://www.jstor.org/ stable/2951599. Borsch-supan, A. & Romer, P. M., 1998. Capital ’ s Contribution and the Nature of Competition. Brookings Papers on Economic Activity. Microeconomics, 1998, pp. 205 – 248. Loo, I. De & Soete, L., 1999. The Impact of Technology on Economic Growth : Some New Ideas and Emprcal Considerations. Lucas, R. E., 1988. On The Machanics of Economic Development. Journal of Monetary Economics, 22, pp. 3 – 42. Mankiw, N. G., Phelps, E. S. & Romer, P. M., 1995. The Growth of Nations. Brookings Papers on Economic Activity, 1995(1), pp. 275 – 326. Romer, P. , 1986. Cake Eating, Chattering, and Jumps: Existence Results for Variational Problems. Econometrica, 54(4), pp. 897 – 908. Romer, P. , 2001. Comment on “ It ’ s Not Factor Accumulation : Stylized Facts and Growth Models ,” by William Easterly and Ross Levine. The World Bank Economic Review, 15(2), pp. 225 – 227. Romer, P. M., 1990a. Are Nonconvexities Important for Understanding Growth ? American Economic Review, 80(2), pp. 97 – 103. 213 Innovative Developments and Economic Growth:The Case of Turkey Romer, P. M., 1990b. Capital, Labor, and Productivity. Brookings Papers on Economic Activity. Microeconomics, 1990, pp. 337 – 367. Romer, P. M., 1990c. Endogenous technological change. Journal of political Economy, 98(5). Available at: http://www.jstor.org/stable/2937632 [Accessed December 28, 2014]. Romer, P. M., 1987. Growth Based on Increasing Returns Due to Specialization. American Economic Review, 77(2), pp. 56 – 62. Romer, P. M., 1986a. Increasing Returns and Long-Run Growth. Journal of Political Economy, 94(5), pp. 1002 – 1037. Romer, P. M., 1994. The Origins of Endogenous Growth. Journal of Economic Perspectives, 8(1), pp. 3 – 22. Available at: http://pubs.aeaweb.org/doi/ abs/10.1257/jep. 8.1.3. Romer, P. M., 2010. What Parts of Globalization Matter for Catch-Up Growth? The American Economic Review, 100(2), pp. 94 – 98. Romer, P. M., 1986b. Why is Japan’s Saving Rate So Apparently High? NBER Macroeconomics Annual, 1(1986), pp. 220 – 233. KEY TERMS Endogenous Growth Higher Education Hi-tech R&D Modelling Patenting Productivity R&D Intensity Technology Transfer QUESTIONS FOR FURTHER STUDY Compare R&D intensity and economic growth rate figures during the last two decades. Describe the inter-relation of patenting and technology export. Explain the relationship between total number of scientific paper publications and economic growth. Design a technology function which has a direct effect on aggregate production. 214 Mehmet Adak EXERCISES What are some of the reasons which led to the Japanese great economic growth that occurred between 1950 and year 2000. How can the Japanese government achieve a high growth rate once more? Clarify the possible reasons of Argentina’s economic decline in the first half of the 20th century. What can Argentina’s cabinet do to improve technological progress. Compare the economies of Ireland and South Korea during the 1990s. Research the facts behind South Korea’s economic growth and Ireland’s economic recession after the year 2000. FURTHER READING Acemoglu, D. Introduction to Modern Economic Growth. Princeton University Press, 2009 Aghion, P. , P. W. Howitt, L. Bursztyn. Economics of Growth. The MIT Press, 2008 Aghion, P. , P. W. Howitt. Endogenous Growth. The MIT Press, 1997 Barro, R. J., Sala-i-Martin, X. I. Economic Growth. The MIT Press, 2003 Jones, C. I., Vollrath, D. Introduction to Economic Growth. 3rd edition W. W. Northon & Company, 2013 Romer, D. Advanced Macroeconomics. 4th edition, Mc Graw-Hill Education, 2011 Wiel, D. N. Economic Growth. Addison Wesley. 2004

###
Chapter Preview

**References**

###
Abstract

The development of information technology has sped up the importance of management information systems, which is an emerging discipline combining various aspects of informatics, information technology and business management. Understanding the impact of information on today’s organisations requires technological and managerial views, which are both offered by management information systems.

This publication takes an interest in the cooperation of business management and management information systems. Its contributions focus on both research areas and practical approaches, in turn showing novelties in the area of enterprise and business management. Main topics covered in this book are technology management, software engineering, knowledge management, innovation management and social media management. This book adopts an international view, combines theory and practice, and is authored for researchers, lecturers, students as well as consultants and practitioners.