Sector dynamics : productivity and economic growth in Europe : a panel data analysis

Publications of the University of Eastern Finland Dissertations in Social Sciences and Business Studies

Tahir Mahmood

Sector Dynamics, Productivity and Economic Growth in Europe:

A Panel Data Analysis

Sector Dynamics, Productivity and

Economic Growth in Europe: A

Panel Data Analysis

Dissertations in Social Sciences and Business Studies No 23

TAHIR MAHMOOD

Sector Dynamics, Productivity and Economic Growth in Europe:

A Panel Data Analysis

Publications of the University of Eastern Finland Dissertations in Social Sciences and Business Studies

No 23

Itä-Suomen yliopisto

Yhteiskuntatieteiden ja kauppatieteiden tiedekunta Joensuu

2011

Kopijyvä Oy Joensuu, 2011 Editor: FT Kimmo Katajala

Sales: University of Eastern Finland Library

ISBN: 978-952-61-0478-2 (print) ISSN: 1798-5749 (print)

ISSN-L: 1798-5749 ISBN: 978-952-61-0479-9 (PDF)

ISSN: 1798-5757 (PDF)

Mahmood, Tahir

Sector Dynamics, Productivity and Economic Growth in Europe: A Panel Data Analysis. 153 p.

University of Eastern Finland

Faculty of Social Sciences and Business Studies, 2011 Publications of the University of Eastern Finland,

Dissertations in Social Sciences and Business Studies, no 23 ISBN: 978-952-61-0478-2 (print)

ISSN: 1798-5749 (print) ISSN-L: 1798-5749

ISBN: 978-952-61-0479-9 (PDF) ISSN: 1798-5757 (PDF)

Dissertation

ABSTRACT

The relationships between economy’s sector share dynamics, productivity and economic growth are increasingly important for Europe. In Europe, the issue of regional competitiveness has taken significant role not only in relation to narrow the gap with the US, but also as part of pursuit of social and economic cohesion.

The objective of this dissertation is to study the relationships between sector level dynamics, productivity, energy intensity, and economic growth in Europe with different panel data methods.

The dissertation gives first an introduction to productivity and growth in Europe. The first article is a panel data inquiry to a long run relationship between sector shares of production and economic growth, i.e. an analysis of structural change in EU region. Article two is a study on labour productivity convergence across 52 industries in Europe. The third article analyzes the growth and trend dependency of energy intensity in Europe 1980-2006 in comparison to some developing countries. Article four investigates the output growth and investment dynamics in Finland at regional level.

Keywords: Sector shares, productivity, economic growth, panel data methods

Mahmood, Tahir

Sector Dynamics, Productivity and Economic Growth in Europe: A Panel Data Analysis. 153 s.

Itä-Suomen yliopisto

Yhteiskuntatieteiden ja kauppatieteiden tiedekunta, 2011 Publications of the University of Eastern Finland,

Dissertations in Social Sciences and Business Studies, no 23 ISBN (nid): 978-952-61-0478-2 (nid)

ISSN (nid.): 1798-5749 (nid) ISSN-L: 1798-5749

ISBN (PDF): 978-952-61-0479-9 (PDF) ISSN (PDF): 1798-5757 (PDF)

Dissertation

ABSTRAKTI

Talouden keskeisten sektoriosuuksien dynamiikan, tuottavuuden ja taloudellisen kasvun suhteiden merkitys kasvaa alati Euroopassa. Alueellinen kilpailukyky Euroopassa on tärkeä, ei yksin kavennettaessa Yhdysvaltojen etumatkaa, mutta myös osana hanketta lisätä Euroopan sosiaalista ja taloudellista yhteenkuuluvuutta. Tämän väitöskirjan tavoitteena on tutkia erilaisten paneeliaineistomenetelmien avulla talouden sektoritasojen dynamiikan, energiaintensiteetin ja taloudellisen kasvun välisiä yhteyksiä.

Väitöskirjassa annetaan aluksi johdanto tuottavuuteen ja kasvuun Euroopassa. Ensimmäinen artikkeli on paneeliaineistotutkimus talouden sektoriosuuksien ja taloudellisen kasvun pitkän aikavälin riippuvuussuhteista, ts. analyysi EU-alueen rakenteellisesta muutoksesta. Artikkeli kaksi on tutkimus tuottavuuden konvergenssista Euroopassa yli 52 toimialan. Kolmas artikkeli analysoi energiaintensiteetin kasvu- ja trendiriippuvuutta Euroopassa vuosina 1980–2006 vertailukohteena joukko vähiten kehittyneitä maita. Artikkeli neljä tutkii tuotannon kasvun ja investointien välistä dynamiikkaa maakuntatasolla Suomessa.

Asiasanat: Sektorien osuudet, tuottavuus, paneeliaineiston analysointimenetelmät

Foreword

I would like to express my deep and sincere gratitude to my supervisor, Mikael Linden, Ph.D., Professor at the Faculty of Social Sciences and Business Studies.

University of Eastern Finland. His wide knowledge and his logical way of thinking have been of great value for me. His understanding, encouraging and personal guidance have provided a good basis for the present dissertation.

I want to express my sincere gratitude to Professor Hannu Piekkola (University of Vaasa) and Professor Juha Junttila (University of Oulu) for their effeorts as the pre-examiners of this Dissertation. Their perceptive comments improved the dissertation considerably. I also thank Hannu Piekkola for serving as the opponent in the public examination.

During this work I have collaborated with many colleagues for whom I have great regard, and I wish to extend my warmest thanks to all those who have helped me with my work at Economics and Business Administration, Joensuu.

On the personal side, I would like to first thank my late parents for making countless sacrifices for my sake. My late mother’s unconditional love has always motivated me to give my best and I thank her for being my inspiration. I owe my loving thanks to my wife Nadia, my son Mohid. They have sacrificed a lot because of my research abroad. Without their encouragement and understanding it would have been impossible for me to finish this work. My special gratitude is due to my brother, my sisters (bhoo) and in-laws for their loving support.

Contents

1 INTRODUCTION ...……….

13

1.1 Background..………13 1.2 Productivity in Main Sectors of Economy and Economic Growth ………15

1.2.1 Productivity of Agriculture, Industrial and Services Sector ……..15

1.2.2 Information and Communication Technologies (ICT) and Productivity………..17

1.2.3 Energy Efficiency and Economic Growth in Europe ………..18

1.3 Productivity and Investment..………20

1.3.1 Economic Growth and Investment ………..20

1.3.2 Productivity and Role of Human Capital ………..20

1.4 Panel Data and Applied Econometric ………21

1.4.1 Random Effect Model (REM) and Fixed Effect Model (FEM) ……22

1.4.2 Bias in the Fixed Effect Model (FEM) and Generalized Method Moments(GMM)………22

1.5 Summary of Articles ……….23

1.5.1 Long Run Relationship between Sector Shares of Production and Economic Growth: A Panel Data Analysis of Structural Changes in the EU Region ………23

1.5.2 Labour Productivity Convergence in 52 Industries: A Panel Data Analysis of Some European Countries ………..24

1.5.3 Trend and Growth Dependence of Energy Intensity in European Economies 1980-2006 ……….25

1.5.4 Output Growth and Investment Dynamics in Finland: A Panel Data Analysis (1975- 2008) ………..26

SOURCES ………

28

ARTICLES ………

33

Article 1

LONG RUN RELATIONSHIP BETWEEN SECTOR SHARES OF PRODUCTION AND ECONOMIC GROWTH: A PANEL DATA

ANALYSIS OF STRUCTURAL CHANGE IN THE EU REGION ...

34

1 Introduction………35

2 Structural Changes and Economic Growth ...37

2.1 Economic Growth and Sector Shares in European Countries ...37

2.2 Relationship between Sector Shares of Production and GDP per Capita ………40

3 Growth Effects of Sector Dynamics ………43

4 Unit Roots and Co-integration……….46

4.1 Panel Unit Root Test………46

4.2 Co-integration between Sector Shares………..48

4.3 Error Correction Models (ECM)………49

5 Industry Share Growth Effects: Granger Non-Causality Tests ………..51

5.1 Fixed Effect Approach ………51

5.2 Results from Granger Non-Causality Test………...52

6 Growth Effects from Equilibrium Sector Share Dynamics ……….54

7 Discussion and Conclusions ………57

Article 2

LABOUR PRODUCTIVITY CONVERGENCE IN 52 INDUSTRIES: A PANEL DATA ANALYSIS OF SOME EUROPEAN COUNTRIES ..

65

1 Introduction………66

2 Labour Productivity and Production Function ………68

3 Industrial Data Base………..71

4 Labour Productivity Convergence ……….73

5 Results ……….74

5.1 Primary Production ………75

5.2 Manufacturing ……….76

5.3 Services ……….77

5.4 Mean Convergence ……….79

5.5 Sectoral Level Convergence ………..79

5.6 β convergence and Labour Utilization ………81

6 Conclusions ………...85

Article 3

TREND AND GROWTH DEPENDENCE OF ENERGY INTENSITY IN EUROPEAN ECONOMIES 1980-2006 ………..

95

1 Introduction ………..96

2 Data and Variables ………...98

3 Economic Growth and Energy Intensity in European Economies ……..100

4 Model and Estimation Procedure ………102

5 Relationship between Economic Growth and De-trended Energy Intensity ………..107

5.1 Rich Countries………107

5.2 Poor Countries ………...110

5.3 Energy Intensity and Main Sectors Contributing to Economic Growth ………112

5.4 Energy Intensity, GDP Per Capita and Population Growth …………..115

6 Conclusions ……….116

Article 4

OUTPUT GROWTH AND INVESTMENT DYNAMICS IN FINLAND: A PANEL DATA ANALYSIS (1975-2008) ………..

124

1 Introduction ………125

2 National and Regional Development in Finland ………..126

3 Variables and Data ……….130

4 Empirical Model for Regional Growth and Investmen ………...131

5 Results ……….132

5.1 Panel Data Analysis of Regional Output Growth, Investment and Employment ………132

5.1.1 Panel Unit Root Tests ………132

5.1.2 Panel Co-integration Test………..133

5.1.3 Panel Error Correction Models (ECM) ………...134

5.1.4 Panel Granger Non-Causality (NGC) Tests ………...136

6 Human Capital and Regional Productivity Convergence ………...139

7 Discussion and Conclusions ……….143

LIST OF TABLES

Article 1 Table 1. Economic Growth in 15 European Countries (Growth Rate of Real GDP per capita) 1970 – 2004 ………..38

Table 2. Unit Root Tests………...46

Table 3. Panel Unit Root Test Results………47

Table 4. Panel Unit Root Test Results on First Differences ………47

Table 5. Panel Co-Integration Results ………49

Table 6. Long Run Parameter Estimates………49

Table 7. Panel ECM Results for Sectoral Shares (N = 15, T =35, 1970-2004) ….51 Table 8. Panel Granger non-causality tests between growth rates of sector shares and GDPc ……….53

Table 9. Panel model results for output growth effects on sector share equilibrium errors. Endogenous variable ……….56

Article 2

Table 5.1. Primary Production………75

Table 5.2. Manufacturing ……….76

Table 5.3. Services……….78

Table 5.4. Industry Productivity Convergence at Sector Level ……….80

Table 5.5 Primary Production (Less Intensive ICT Using Industries) ……….82

Table 5.6. Manufacturing ………82

Table 5.7 Services ……….83

Article 3 Table 1. Economic Growth and Energy Intensity in Europe 1980 – 2006 …..101

Table 2. Results from the Estimation of Model 2………...105

Table 3. Panel unit root test results ………107

Table 4 Panel Results for Model M1 and M2 (N = 19, T =26, 1980-2006) …..109

Table 5 . Results from the Estimation of Equation (1) ……….110

Table 6. Panel Results for models M1 and M2 (N = 10, T = 26, 1980-2006) ..111

Table 7. Sector Output Growth and De-trended Energy Intensity …………..114

Table 8. GDP per capita growth, Population growth, and De-trended ……116

Table A1 Panel Unit Root Test Results for Poor countries ………122

Article 4 Table 1. Panel unit root test results ……….132

Table 2. Panel unit root test results on first differences ………...133

Table 3. Panel Unit Root Test ……….133

Table 4 . Co-Integration Relation at Panel Level……….134

Table 5. Panel Results (N = 20, T =33, 1975-2007) ………135

Table 6. Panel Results (N = 20, T = 33, 1975-2007) ………..137

Table 7. Region Time Series Granger non-Causality Tests ………..138

Table 8. Panel Results (N = 10, T = 25, 1981-2005) ……….142

Table A3. Results for Output AR(2) Model ………150

Table A4.1. Time Series Co-integration Relation at University Region …….152

Table A4.2. Panel Co-integration Relation ………..153

LIST OF FIGURES

Article 1 Figure 1. Agriculture Shares of GDP (% of GDP) for 15 European Countries in 1970-2004 ………..38

Figure 2. Industry shares of GDP (% of GDP) for 15 European countries in 1970-2004 ……… 39

Figure 3. Services Shares of GDP (% of GDP) for 15 European Countries in 1970-2004 ………...40

Figure 4. Relationship between Real GDPc and Sector Shares in 15

European Countries 1970-2004……….42 Figure 5. Relationship between Real GDPc growthand Sector Shares in 15 European Countries 1970-2004 ……….42 Article 3

Figure 1: Time cross plot of GDP –levels and Energy/GDP –ratio …………..100 Figure 2: Time cross plot of Energy/GDP -ratio and dlnGDP ………..102 Figure 3. Trends in Energy Intensity in European Countries ………..104 Figure 4. De-trended Energy Intensities in European Countries ………106 Figure A1. Upward trend in Energy Intensity of Poor Countries …………...121 Figure A2: De-trended Energy Intensity of Poor Countries ………122 Figure A3: Time cross plots of GDP and Energy/GDP of Poor Countries ….123 Figure. A4: Time cross plots of Population and Energy/GDP of

Poor Countries ………123

Article 4

Figure 1. Regional Real Output Growth and Real Investment

1975- 2008 ………128 Figure 2. Regional Real Output Growth and Number of Wage Earners

1975- 2008 ………129 Figure 3. Regional Real Output Growth and Number of Entrepreneurs

1975-2008 ………..130

1 Introduction

1.1 BACKGROUND

Today, in the period of globalization and of broad single currency, European region must closely concern the competiveness of its production system. No spontaneous adjustment mechanism is at work to counterbalance insufficiencies in economic growth and lack of productivity (see, e.g. Capello 2008).

Productivity is the cornerstone of economic growth. The Europeans are richer than the average person in the Third World primarily because Europe is more productive. Productivity also affects region’s competitive position in the world market. In short, productivity is a source of the high standard of living in the European economies.

Roughly productivity is a measure of output of goods and services per unit of input, for example, per unit of labour (i.e. labour productivity, see Fernando and Yvonn 2008), and per unit of capital (i.e. capital productivity, see Solow et al.

1996). Like labour or capital productivity, energy productivity measures the output and quality of goods and services generated with a given set of energy inputs (see Hartmann et al. 2008). Finally, productivity can be an output of goods and services per unit of all production resources. Their different shares in aggregate production reflect the structural composition and change of economies.

Countries are not equally endowed with natural resources. For example, some countries benefit from fertile agricultural soils, while others have to put a lot of effort into artificial soil amelioration. Some countries have discovered rich oil and gas deposits within their territories, while others have to import them. In the past a lack or wealth of natural resources made a big difference in countries' development. Today the wealth of natural resources is not the most important determinant of development success. Consider high-income countries. Their high economic development allows them to use their limited natural wealth much more productively (efficiently) than is possible for many less developed countries. An extensive literature which is comparing the development of different countries shows that the efficient use of productive resources like physical capital, human resources, and natural resources are widely recognized as the main indicator of country’s or region’s level of economic development.

However such analyses are extremely challenging, primarily because of the difficulty of measuring values on elements of natural and human capital.

In the factor endowment model of neoclassical growth, income differences between countries are due to different capital-labor ratios. The Balanced growth models (BGM) are based on the Kaldorian facts imply that in the long run capital-labour ratio is roughly constant over time. However, “Kuznets facts”

refers to reallocation process taking place in the economy’s sectoral shares during its development. This structural change entails that income share of agriculture declines and share of services increase in the economy. These sectoral dynamics are associated with the rise in per capita income. A study by Kongsamut et al. (2001) proposed a generalized growth path model where in balanced growth is consistent with the dynamics of structural change. However the result is an outcome of quite demanding restriction on sector endowments.

The growth performance of European Union has been elaborated and how it has been undergone transitions during the second half of the 1990,s (O’Mahony and Van Ark, 2003). The average annual real GDP growth of EU-15 remained constant at 2.2 per cent, but the labour productivity growth slowed dramatically. This structural slowdown for European economies is captured by extensive literature that focused on sectoral dynamics that could influence the slowdown in productivity growth in European economies (see Dew-Becker and Gordon 2006, Bourles and Cette 2007, and Jimeno, Moral and Saiz 2006). The decline in the structurally stable productivity growth rate exacerbated the slowdown in productivity growth in Europe, which had started in 1995 (see Van Ark and Inklaar (2005).

In this dissertation we use different approach in order to analyze the productivity and economic growth of European economies. We apply different panel data models to analyze the long run phenomena of productivity and economic growth. We also use some novel approaches to analyze the relationship between productivity and economic growth for the panel of some European countries. For example, in first article we use GDP¹ per capita and shares of agriculture, industrial and services sectors to analyze long run phenomena. In second article, we use productivity as an output of goods and services per unit of labour and per unit of work hours respectively. Here, we study at disaggregated level of 52 industries, which is novel approach. In third article, we use de-trend energy intensity as a measure of the energy efficiency (i.e. Productivity). This is also novel approach. In fourth article, we compare the

1 Here, Gross Domestic Product (GDP) is a value of all final goods and services produced in a country in one year. GDP can be measured by adding up all economy's incomes, i.e. wages, interest, profits, and rents or expenditures that are consumption, investment, government purchases, and net exports (exports minus imports). Both results should be the same because one person's expenditure is always another person's income, so the sum of all incomes must equal the sum of all expenditures.

We used World Bank national accounts data. According to International Standard Industrial Classification (ISIC) given in World Bank national accounts data, GDP per capita is the sum of gross value added by all resident producers in the economy plus any product taxes and minus any subsidies not included in the value of the products. It is calculated without making deductions for depreciation of fabricated assets or for depletion and degradation of natural resources.

regional productivity in Finland by using their regional values for capital, human capital and output per capita.

This dissertation contains four articles. Their contents are following.

Article 1. Long run relationship between sector shares of production and economic growth: A panel data analysis of structural change in the EU region.

Article 2. Labour productivity convergence in 52 industries: A panel data analysis of some European countries.

Article 3. Trend and growth dependence of energy intensity in European economies 1980-2006.

Article 4. Output growth and investment dynamic in Finland: A panel data analysis (1975-2008).

The outline of the dissertation is as follows. Section 1.2 is an introduction that discusses productivity and economic growth in general. Here, the special emphasis is laid on the literature on the productivity of different sectors (i.e.

agriculture sector, industrial sector, services sector, ICT- sector, energy sector), and economic growth. Section 1.3 reviews the role of investment and human capital in context of regional productivity and output growth. Section 1.4 focuses on the methods used in the dissertation. Finally section 1.5 gives the summaries of the dissertation articles.

1.2 PRODUCTIVITY IN MAIN SECTORS OF ECONOMY AND ECONOMIC GROWTH

1.2.1 Productivity of Agriculture, Industrial and Services Sector

An important insight of classical development economics was that economic growth is intrinsically linked to changes in the structure of the production.

According to this view, industrialization is the main source of technical change, and therefore, overall productivity increase is mainly a result of the reallocation of labour from low to high productivity sectors. Initially, agriculture is a developing economy’s most important sector. As income per capita rises, agriculture loses its primacy giving a way first to a rise in the industrial sector and then to a rise in the service sector. These two consecutive shifts are called industrialization and post-industrialization (or “de-industrialization”). As the citizens’ incomes increase, they start to demand also non-agricultural products.

At same time, because of new farm techniques and machinery, labour productivity increases faster in agriculture than in industry. This makes agriculture products relatively less expensive. This further diminishes their share in gross domestic product (GDP). The same trend in relative labour productivity also diminishes the need for agriculture workers, while

employment opportunities in industry grow. As a result industrial output takes over a larger share of GDP than agriculture (see Taytyana et al. 2000).

The industries (goods producing sectors) and services sectors are known as engines of the development. Developed economies heavily relay on big industrial and efficient services sectors. Hence, a debate exits why there are structural changes in the developed economies in the past few decades. The neoclassical approach is based on the view that structural change is an unimportant side effect of the economic development (see Echavarria 1997). On other side economist associated with the World Bank found that growth is brought by the changes in sectoral composition of the economy (Baumol et al.

1989).

The empirical research on the impact of industrial development on economic growth started with Kaldor (1966). The relation of industrial sector with economic growth has it roots in Kaldor views about manufacturing sector.

Kaldor (1966) argued that the industrial sector is the “engine of growth”. Kaldor explained his ideas by giving three laws. According to Kaldor’s first law the faster the rate of growth in manufacturing sector the faster the growth of overall gross domestic product. He argued that when manufacturing sectors develops then other sectors of the economy also develops through spill-over effects. The second law, which is also known as Verdoorn’s law (1949), states that there is strong relationship between the growth of labour productivity in the manufacturing sector and the growth of output in manufacturing sector. In third law Kaldor states that productivity growth is positively related with the employment in the manufacturing sector where as it is negatively related with non-manufacturing sectors.

Developed economies have undertaken a process of industrial transformation since 1920’s. This industrial transformation and structural change increased the importance of services sector. Service industries in Europe have shown remarkable dynamism, which has enhanced their possible role as the new engine of growth in knowledge-based economy. Services are becoming a key engine of growth, first and foremost, because of the high technological content and great knowledge intensity that characterize their production and provision (Evangelista 2000, Drejer 2004, Hartwing 2008).

Typically some sectors contract and some expand making their growth effects unpredictable as may feedback and spill-over effects are evident among the sectors. For, example, the expansion of the service sector relative to the rest of the economy leads to a reduction in the long run rate of growth of output per capita (see for example Baumol et at. 1985).This could be due to the fact that services are mostly non physical production. Baumol (1967) argued that scope of productivity growth in the services sector is slower than in the sectors that produce goods. In this sense Baumol’s approach is important as it argues that stagnant growth is possible and almost evident, if the productivity growth differs across the economy’s sectors. Therefore, some empirical questions are

still open. The first question is “Do the sector shares in economies adjust to each other along to long run stationary path?”, and the second question is “How the long run sector adjustments are related to GDP growth across the countries?”.

The first article of this study addresses these two questions in details.

1.2.2 Information and Communication Technologies (ICT) and Productivity

In concentration of ICT there is extensive literature on intangible investment (Piekkola 2010, Ilmakunnas and Piekkola 2010). A study by Gorzing, Piekkola and Riley (2011) developed a methodology for evaluating the investment of companies in intangible assets where firms produce three types of goods. First type of good is information and communications technology (ICT), second research and development (R&D), and third type is organizational capital (OC).

Study argued that the total shares of intangible capital type workers are typically around 18% of all workers. The ICT capital in the form of software and database is one of the few items recorded in national accounts, while R&D investment is not currently recorded as part of GDP. A study by Corrado, Hulten, and Sichel (2005) found that the business fixed investment in intangible assets may have been large as the spending on tangible capital. The study concluded that the inclusion of un- recognized business intangible capital in national accounts could alter the average growth rate of real output and labour productivity in the late 1990,s. Therefore, role of ICT related industries in economic growth is very important.

The existence of information and communication technologies (ICT) has potential to enhance the productivity in many sectors of economy. Many services industries, due to the intangible and knowledge based nature of the activities they carry out, are closely related to the core of new general purpose technologies, since they are active producers and users of ICT (Van Ark et al., 2008). The adoption and use of ICT related innovations create new opportunities for knowledge exchanges between services and manufacturing industries (e.g.

software, hardware and telecommunications). Therefore, linkages between these interrelated branches of economy are increasingly becoming a key factor of economic growth and competitiveness (Guerrieri and Meliciani, 2005).

Information and communication technologies (ICT) affects economic growth both as a component of aggregate output in the form of ICT production and as component of aggregate input in the form of ICT capital services. One of the main sources of labour productivity growth is ICT capital deepening. The share weight of ICT capital services per hour worked increases. A study by OECD indicates that the contribution of ICT to OECD economies is significant for productivity performance. Estimates show that the contribution of ICT accounted for between 0.3 and 0.8 percentage points of growth in GDP and labour productivity over period 1995-2001 (OECD, 2002). For example, ICT’s average contribution to French GDP growth was estimated to be approximately

0.2% per year over 1969 and 1999. This figure increased to 0.3% between 1995 and 1999 (Cette et al. 2001). Similarly the contribution of ICT to US labor productivity shows that the ICT-related industries are indeed driving the U.S.

productivity (see Stiroh 2002).

One of the major achievements of develop economies in the past decade is a revival of labour productivity. One of the reasons for this positive development is linked to the effective use and production of ICT (see OECD 2002). The channels through which ICT affects labour productivity are numerous. First, ICT as an input is considered to increase the productivity of not only labour productivity but also the productivity of the non-ICT capital.

Second, through their networking effect, ICT significantly reduce transaction costs for firms and hence help to improve overall efficiency in the economy.

However, studies also indicate that most advanced European economies are lagging behind US and other emerging economies’ productivities. This can also attributed to insufficient level of investment and heterogeneous policy environment across European economies. But two questions related to ICT raises here. The first question is “Are the ICT-related industries (i.e. ICT using and ICT producing industries) contributing to labour productivity convergence in Europe like in US?” The second is “What is the speed of convergence in all industries?” The second article of the dissertation answers to these questions in details.

1.2.3 Energy Efficiency and Economic Growth in Europe

Mainstream economists think that the capital, labour, and land as the primary factors of production, while such goods as a fuels is intermediate input. The prices paid for all the different inputs are seen as eventually being payments to the owners of the primary inputs for the services provided directly or embodied in the produced intermediate inputs (Stern 2003). This approach gives more focus to the primary inputs, and in particular, to capital and land, and gives a much lesser value to the role of energy in the growth process. The primary energy inputs are stock resources such as oil deposits. These are not given an explicit role in the standard growth theories. Therefore, the ideas about the role of energy in the mainstream theory of growth tend to be fairly convoluted.

However, capital, labour, and in the longer term even natural resources, are reproducible factors of production, while energy is not a reproducible factor of production while of course energy vectors (fuels) are (Stern 2003). Therefore, scientists in natural sciences and some ecological economists have placed a very heavy emphasis on the role of energy and its availability in the economic production and growth processes. Analysis of energy productivity provides a framework for understanding the relationship between energy demand (i.e. use) and economic growth. Higher energy productivity can be achieved by higher energy efficiency that reduces the energy consumed to produce the same level of energy services (e.g. a more efficient motor engine produces the same output for

less energy input). The more efficient use of energy is more desirable than seeking to reduce end-use demand by compromising economic growth.

A global energy demand is expected to rise by almost 35 percent until 2030. The increasing global GDP and population with ever increasing expectations of wealth and lifestyle are likely to mean a rapid expansion in energy supplies of all types. To put this into perspective, China’s GDP will likely be larger than the USA or Europe GDP by 2040. Therefore, it is understandable to study the relationship between energy and economic growth in Europe.

With current policies, European energy demand will grow yearly at 1.2 percent to 2020. Europe represents 17 percent of global energy consumption, less than the 22 percent of the United States, the world’s largest energy consumer, but more than China, which is at 14 percent. Over the period 1990-2002, European GDP grew at an annual average rate of 2.2 % and total energy consumption at annual average rate of 0.5%. As a result, total energy intensity in the EU fall as at the average rate of 1.7% (see EEA 2008). Energy intensity is a measure of the amount of energy it takes to produce a dollar's worth of economic output.

In the EU-15 during the early 1990s, a combination of low growth in GDP, low fossil fuel prices (see, EN31), and a general low priority for energy saving in the most member states contributed to a slowdown in the reduction in final energy consumption intensity. Since then energy-efficiency improvements have become more important. Recently published results indicate that most manufacturing industries (except textiles) experienced increasing energy productivity between 1990 and 2002 in the EU-15, influenced by improved production processes and innovative technologies (SAVE 2003). Sweeping improvement in the energy productivity of European economies could prevent the runway energy demand and consumption. McKinney report says that Europe has an opportunity to increase energy productivity that would halt energy demand growth in the region (see, Hartmann et al. 2008).

Therefore, we can reduce our energy consumption through boosting energy productivity. This implies that for the developed economies a negative relationship between energy intensity and economic growth is valid. Thus the

“energy-saving” high GDP-level effect is extended to be valid also for economic growth. Main argument is the fact that service economy is less energy demanding than preceding economic epochs, and energy saving economy is only possible with high technology (Stern 2003). However, if the technology and energy composition effects (i.e. declining trend of energy/GDP ratio) are removed, the growth effects may still be energy saving. This hypothesis has not yet been formally tested. Therefore, in third Article of this dissertation, we test this hypothesis.

1.3 PRODUCTIVITY AND INVESTMENT

1.3.1 Economic Growth and Investment

The notion of growth as increased stocks of capital goods was stressed by the Solow-Swan growth models. These focused on the relationship between labour- time input, capital goods input, output, and savings. However, for a new long run steady state, the role of technological changes became crucial, even more important than the accumulation of capital. Models assume that countries use their resources efficiently and that there are diminishing returns to capital and labour increases. From these two premises, the neoclassical model makes three important predictions. First, increasing capital relative to labour creates economic growth, since people can be more productive given more capital.

Second, poor countries with less capital per person will grow faster because each unit of investment in capital will produce a higher return than rich countries with ample capital. Third, because of diminishing returns to capital, economies will eventually reach a point at which no new increase in capital will create economic growth. This point is called a "steady state" (Solow 1957).

However modern economic research shows that the baseline version of the neoclassical model of economic growth is not supported by the empirical evidence. Calculations made by Solow claimed that the majority of economic growth was due to technological progress rather than inputs of capital and labour. However, calculations made to support this claim are invalid as they do not take into account changes in both investment and labour inputs (Jorgenson 1988, 1990). Landes’ (1969) statement that “the machine is at the heart of the new economic civilization” is typical of accounts that have assigned a central role to mechanization. Technology embodied in machinery has been, as Mokyr (1990) says, “the lever of riches”. Work in the growth accounting tradition of Solow has typically concluded that capital accumulation accounts for only a relatively small fraction of productivity growth (e.g. Denison 1967, Denison and Chung 1976). However, Jorgenson’s more sophisticated and much more disaggregated growth accounting exercises find substantial complementarily between equipment investment and total factor productivity growth, and thus a somewhat larger role for investment in enabling productivity growth.

Many other studies also present investments as an important determinant of long run economic growth (e.g. see, De Long and summers 1991, 1992). The results in fourth Article of the thesis are also reconcilable with the capital fundamentalist view. The rate of physical capital formation in Finland is also the primary engine of long run growth.

1.3.2 Productivity and Role of Human Capital

The main concept of human capital literature is anticipated by Friedman and Kuznets (1945). Mincer (1958) conducted the study on the return on schooling which was the first formal analysis (see also, Becker 1964). Over the last decades,

the importance of human capital has taken central role in discussion regarding the growth because developed economies have increasingly evolved towards what has been called “the knowledge based economy” (OECD 2005).

Human capital is seen today to be a crucial feature of economic growth.

However, links between human capital and economy development may not necessarily be the same as those between human capital and regional development. One of the reasons is that the human capital in a region has an impact on the aggregate productivity in the economy via the externalities associated with it (see, Faggian and McCann 2009). Therefore, in the terms of regional development issues, the role played by tertiary education is important, rather than primary or secondary education (Faggian and McCann 2006). This has led to a focused analysis on the interaction between higher education institutions and the regional productivity. Therefore, in order to understand the link between human capital and regional productivity we must consider the role of higher education. In article 4 of dissertation, we try to incorporate the human capital aspects into analysis, and ask “Do the de-centralized higher education system affect the regional productivity convergence?"

1.4 PANEL DATA AND APPLIED ECONOMETRICS

Panel data analysis endows regression analysis with both spatial and temporal dimension. The spatial dimension pertains to a set of cross-sectional units of observation. These could be countries, states, counties, firms, commodities, groups of people, or even individuals. The temporal dimension pertains to periodic observations of a set of variables characterizing these cross-sectional units over a particular time span. An example of a panel data set is a collection of 15 countries for which there are the same economic variables—such as labour productivity, GDP growth, GDP per capita, and employment—collected annually for 30 years. This pooled data set, sometimes called time series cross- sectional data, contains a total of 15 30 = 450 observations.

The panel data has the merits of using information concerning cross section and time series analyses. It can also take heterogeneity of each cross- section unit (e.g. regional differences) explicitly into account by allowing for individual specific effects (see, Davidson and MacKinnon 2004). Therefore, it gives more variability, less collinearty among variables, more degree of freedom and more efficiency (see Baltagi 2001). The objective of most empirical studies in economics is to determine whether a change in one variable, say

x

, causes a change in another variable, say

y

. Goldberger (1972) defines a structural model as one repressing a causal relationship, as opposed to a relationship that simply captures statistical associations. A structural equation can be obtained from a formal economic model, or it can be obtained through informal reasoning. Some time the structural model is directly estimable. However, typically we must

combine auxiliary assumptions on other variables to arrive at an estimate able model. The error term

u

can consist of a variety of things, including omitted variables. Whether this is the case depends on the application and model assumptions made. Therefore, it is very important to obtain consistent estimators in the presences of omitted variables with panel data.

1.4.1 Random Effect Models (REM) and Fixed Effect Models (FEM)

Let

y

and

x

be observable random variables, and let

c

be an unobservable random variable. The vector (

y x x , ,

_{1 2}

,..., x c

_k

, )

represents the population of interest. As is often the case in applied econometric, we are interested in partial effects of the observable explanatory variables

x

_j in the population regression function (see, Chamberlain 1982). In panel data, the basic unobserved effect model can be written, for a randomly drawn cross section observation

i

, as (1)

y

_it

x

_it

c

_i

u

_it where

t 1 , 2 ,...., T

In modern econometric parlance, “random effect” is synonymous with zero correlation between observed explanatory variables and the unobserved effect:

( , )

_{it i}

0, 1, 2,...,

Cov x c t T

. In applied papers, when

c

_i is referred to as an individual random effect, then

c

_i is assumed to be uncorrelated with

x

_it. In the traditional approach to panel data models

c

_iis called a random effect (RE) when it is treated as a random variable and a fixed effect (FE) when it is treated as a parameter to be estimated for each cross section observation. In this thesis we use fixed effect approach to handle the cross-section differences.

Note that in FE-models individual effects (cross-section dummies) can correlate with

x

_it. Hausman test can be used to determine if the RE-approach is applicable.

1.4.2 Bias in the Fixed Effect Model (FEM) and Generalized Method of Moments (GMM)

In the past, researchers have regarded estimated fixed effects as “nuisance”

parameters that cause efficient loss and bias in the estimation. Thus difference models are recommended although FE-parameters often convey useful information in industrial, labour, environmental and health economics (e.g. see, McClellan and Staiger 2000, Murdock 2006). Biases are most acute when N is large compared to T.

The estimation of panel data model with lagged dependent variable in the set of regressors also produces biased coefficient estimates. The basic problem of using OLS is that the lagged dependent variable is correlated with the error term as the dependent variable

y

_it is a function of _ít

c

_i

u

_it, and

it immediately follows that

y

_{it 1} is also a function of _ít. Note that the fixed effect (FE) estimators are also biased and inconsistent unless the number of time periods is large (for details, see e.g. Baltagi 2001). In such cases, the instrumental variable (IV) estimator (Anderson and Hsiao 1981), and generalized method of moments (GMM) estimator (Arellano and Bond 1991) are both widely used. In this dissertation we exploit the GMM-DIFF procedure of Arellano and Bond, which suggests first to take differences of the variables in the model and then to use lags of the dependent and explanatory variables as instruments for the lagged dependent.

1.5 SUMMARY OF ARTICLES

1.5.1 Long Run Relationship between Sector Shares of Production and Economic Growth: A Panel Data Analysis of Structural Change in the EU Region.

The growth process in the European countries entailed a dramatic change in the employment structure, involving a shift from the primary sectors into industry and, subsequently, into services (see, Maddison 2001). The positive association between economic growth and the share of services has been documented by a number of studies including Fisher (1935), Clark (1941), Kuznets (1957), Chancey (1979), and Fuchs (1980). Clark traced the observation of this relationship back to Sir William Petty and proposed that the shift of the working population from agriculture to manufactures and from manufactures to services in the course of economic growth can be called as Petty's Law.

Theoretically there exist conflicting arguments how these sectors are related to economic growth and development, i.e. how the shares of three major sectors develop in time. Thus, shifts of resources, output and employment between different sectors accompanying the process of economic growth are recognized as a possible challenge for adjustment in industrial economics. The mainstream hypothesis in economics tends to classify this as a short term problem of adjustment. The challenge of combining a model of stable growth along a steady path with structural change between sectors with different productivity paths was formulated by Baumol (1967). In this model the service sector as the stagnant sector with low productivity growth attracts labour and thereby lowers the overall growth rate of the economy (Kratena 2005). It has been shown (e.g. Echavarria 1997, Kongsamut, Rebelo and Xie 2001, Bonatti and Felice 2008) that whether Baumol’s pessimistic outcome is reproduced or not depends on the functional forms of the utility function of consumers, and on the differences in technological progress between the sectors. However, typically the structural changes ceases when the stable path is reached. Baumol’s result changes also considerable by taking into account intermediate demand for

services. In case the small productivity increases in the services sectors are not a threat for an overall stable rate of growth (Oulton 2001).

With respect to arguments above we argue that the main empirical question is whether sector adjustments and their growth impacts are equilibrium phenomena or not. We focus on two questions: 1) Do the sector shares in economies adjust to each other along long run stationary paths, and 2) How the long run sector adjustments are related to GDP growth across the countries. These questions are analyzed with data from 15 European countries (the Schengen countries) in period 1970-2004. We estimate co-integration and error correction (EC) models, and conduct Granger non-causality (GC) tests over the panel of Schengen countries with respect to sector growth patterns and GDP per capita growth. The results indicate that equilibrium forces are found between pair-wise sector dynamics but their economic growth effects are different. Likewise the dynamic effects between growth rates of sectors shares and GDP per capita growth rate are varying. However some typical balanced growth model implications are not rejected since the analyzed countries have similar sectoral structural change.

1.5.2 Labour Productivity Convergence in 52 Industries: A Panel Data Analysis of Some European Countries

The lower productivity performance in Europe as compare to US has caused some concern about the income growth prospects in Europe. The lack of productivity growth has been seen as the culprit of the sluggish economic growth that Europe has experienced in the last few decades. Strategies to overcome it have been on the agenda of the European Union for a long time (European Commission 2003, Sapir 2004). Although some convergence has been seen in Europe, there are still national economies exhibiting regional inequalities casting doubt on the relevance of the national level accounting for the dispersion of productivity performance. Some studies have argued that the European integration process has favored specialization and convergence of regions across national borders rather than uniform geographic convergence (Quah 1997, Fatás 1997). The empirical literature shows that income differentials across United States are narrower than in Europe (Puga 1999).

Since the mid-1990’s, the Nordic EU countries particularly Sweden and Finland, have experienced stronger hourly labour productivity growth than the larger Euro-area countries. Combined with a high level of labour utilization, this has resulted in a “structural” labour productivity level that is relatively high compared with some of those larger Euro-area countries. Innovation and technological changes have played a major role in raising labor productivity growth in the Nordic EU countries industries (see, Annenkov et al. 2005).

However the case for rest of Europe is not so clear. Industry level differences in labour productivity are large and the structural change among and between the industries varies. The adaption of ICT and its productivity gains are still largely

unknown across the European industries. Similarly the industry level convergence of labour productivity as a result of EU common market area and the unified technology policy needs a closer look.

The second article analyzes the β-convergence and speed of convergence for the European industries. We use cross country fixed effect estimation method for panel data of disaggregated level of 52 industries for 13 European countries in period 1979 - 2003. The analysis focuses also on economy sector level, labour utilization, and ICT productivity effects. In agriculture sector and in service sector, the existence of β-convergence is found for all industries. In manufacturing sector, convergence is found for all industries except for electronic and computing equipment industries. In general the speed of convergence estimates show slow adjustment. Speed is highest in the capital intensive industries. In primary production the convergence is slowest in agriculture and fastest in fishing industry. The convergence speed is fastest in oil refining and nuclear fuel manufacturing industries. By augmenting the productivity models with labour utilization variable speeds up the convergence.

Labour utilization is positive related to productivity growth in primary production industries, ICT producing manufacturing industries, and ICT producing services industries. Results indicate that the European industries are still far behind the US state level industry convergence.

1.5.3 Trend and Growth Dependence of Energy Intensity in European Economies 1980- 2006

Energy intensity is the ratio of energy consumption to GDP. It is a measure of the energy efficiency of the economy. The rising energy prices and high GDP- level drive energy intensities down. Decreasing energy intensity in the developed countries is a largely accepted hypothesis in the energy literature.

However the relationship between energy intensity and economic growth is not so clear.

Energy intensity is an important determinant of carbon emission since energy combustion is responsible for roughly 98 percent of carbon emission (see EEA 2007). The reduction in energy demand can be effected by more efficient energy use in production and consumption. Thus an important question is if the improvement in energy intensities is only a result of efficiency gains of technology trend found in the high income countries. Thus, are some GDP growth effects in energy intensities not yet observed?

The paper contributes to literature by under taking an econometric analysis of an intensity trends at each cross section. A new approach based on de-trended energy intensity is introduced into growth analysis. The dependency of de-trended energy intensity on GDP growth and its main contributing sectors (i.e. industry, services and agriculture) is studied in details. As part of our analysis we discover key determinants of changes in energy intensity. For the high income group (i.e. European countries) de-trended energy intensities react

negatively to GDP growth. Thus economic growth, not only high GDP level, decreases energy intensity in Europe. Hence, the role of oil price shocks and the business cycles on the falling energy intestines in the European economies are questioned. The impacts of growth in sector outputs, GDP per capita growth, and population growth on de-trended energy intensity are also large. Contrary to this, in poor countries both GDP and population growth requires intensive energy use, and the business cycles and the oil price shocks affect energy use positively.

1.5.4 Output Growth and Investment Dynamics in Finland: A Panel Data Analysis (1975- 2008)

Economic growth corresponds to a process of continual rapid replacement and reorganization of human activities facilitated by investment in physical capital motivated to maximize returns. Investment is the amount of new capital that is added to the existing capital stock in a given year. Economists refer to investment as a flow variable, and to the capital stock as a stock variable. We can think of capital as basically being the equipment, structures and inventories that help improve the productive capability of the economy. In short the capital stock of a country is simply the quantity of productive assets that produce goods and services.

The Keynesian approach to growth stresses the significance of investments in the growth process, i.e. the multiplier effect. This was augmented with accelerating principle that considers the opposite causation, i.e. how the growth rate of economy spurs the investments. A large literature has found a robust positive relationship between fixed investment and long run economic growth (Levine and Renelt 1992, Mankiw et al. 1992, De Long and Summers 1991, 1992).

A faster growth rate is trigged by the higher investment rates. On other hand recent empirical evidence contrasts this view, and suggests that the causality links runs in opposite direction. A higher economic growth leads to higher investment rates. Studies by Blomström et al. (1996) and Caroll and Weil (1994) found that growth rates Granger-cause investment rates, but investment rates do not Granger cause growth rates.

We analyze the growth investment -relationship with the regional data from Finland. The growth experience of different regions (provinces and districts) is well documented but the statistically sound analysis stressing the role of investments is hard to find. First, we used two-way FE error correction model (ECM) in order to analyze the long run relationship between output growth and investment. Second, we conduct Granger non-causality test both on the regional level and on the panel form. Results show that a positive association between growth rates of output and investment is found, and the investments predict output growth.

The analysis is widened to relationship between labour productivity measured as value added per employment and human capital. Here, we try to

incorporate the human capital aspects into analysis, and ask do the de- centralized higher education system affect the regional productivity convergence in Finland. The growth literature of developed economies and new growth theories stress external economies as main cause of growth (Chenery et al. 1986, Romer 1986). Therefore, human capital is an important input in regional development and growth. Results show that the growth of human capital per capita is positively related with regional output growth per capita in Finland but it is not statistically significant. However the conditional productivity convergence in Finland was not rejected.

Sources

LITERARTURE

Anderson, T.W. and Hsiao, C. (1981): Estimation of Dynamic Models with Error Compo-nents, Journal of the American Statistical Association, 76, 598-606.

Annenkov, A., Madaschi. C (2005): Labour Productivity in the Nordic EU Countries, A Comperative Overview and Explanatory Facts, European Central Bank Occasional Paper, No.39.

Arellano, M. and Bond, S. R. (1991): Some Tests of Specification for Panel Data:

Monte Carlo Evidence and an Application to Employment Equations, Review of Economic Studies, 58, 277-97.

Ark, Van B., M.O, Mahony and M.P. Timmer. (2008): European Growth: the End of Convergence, Journal of Economic Perspectives, 22, 22-44.

Baltagi, B. H. (2001): Econometric Analysis of Panel Data (2^nd ed.). New York: John Wiley and Sons.

Baumol, W. J. (1967): Macroeconomics of Unbalanced Growth: The Anatomy of Urban Crisis, American Economic Review, 57, 415-426.

Baumol, W. J., Batey, B. S. and E. N. Wolff (1985): Unbalanced Growth Revisited:

Asymptotic Stagnancy and New Evidence, American Economic Review, 75, 806-17.

Baumol, W. J., Blakman, B.J., and E. N. Wolff. (1989): Productivity and American Leadership. Cambridge UP.

Blomstrom, M., Lipsey, R. E. and Zejan, M. (1996): Is Fixed Investment the Key to Economic Growth?, Quarterly Journal of Economics, 111, 269-76.

Bonatti, L. and Felice, G. (2008): Endogenous Growth and Changing Sectoral Composition in Advanced Economics, Structural Change and Economic Dynamics, 19, 109-131.

Bourles, R. and Cette, G. (2007): Trends in Structural Productivity Level in the Major Industrialised Countries, Economic Letters, 10, 10-16.

Capello. R. (2008): Regional Economies in its Fifties: Recent Theortical Directions and Future Chalanges, Annals of Regional Science, 42, 747-67.Clark, C.

(194 1). The Conditions of Economic Progress, London: Macmillan.

Carroll, C. D. and Weil, D. N. (1994): Savings and Growth: A Reinterpretation, Carnegie-Rochester Conference Series on Public Policy, 40, 133-92.

Cette, G., Mairesse, J. and Kocoglu, Y. (2001): The Diffusion of ICT,s and Growth of the French Economy Over the Long-term, 1980-2000, International Productivity Monitor, No. 4, Spring.

Chamberlain, G. (1982): Multivariate Regression Models for Panel Data, Journal of Econometrics, 18, 5-46.

Chenery, H.(1986): Industrialization and Growth: A Comparative Study. Oxford:

Oxford University Press.

Corrado, C. Hulten, C. and Sichel. D. (2005): Measuring Capital and Technology:

An Expanded Framework, University of Chicago PRESS, 0-226-116123.

Davidson, R., and MacKinnon, J.G. (2004): Econometrics Theory and Methods, New York: Oxford University Press.

Denison, Edward F (1967): Why Growth Rate Differ. Postwar Experience in Nine Western Countries, Washington, DC. The Brookings Institution.

Denison, Edward F., and William Chung (1976): How Japan’s Economy Grew So Fast. Washington, DC: The Brookings Institution.

De Long, J. B. and summers, L. H. (1991): Equipment Investment and Economic Growth, Quarterly Journal of Economics, 106, 445-502.

De Long, J. B. and summers, L. H. (1992): Equipment Investment and Economic Growth: How Strong is the Nexus?, Brookings Papers on Economics Activity, 157-211.

Derjer, I. (2004): Identifying Innovation in Surveys of Services, a Schumpeterian Prespective. Research Policy 33, 551-562.

Dew-Becker, I. and Gordon, R. (2006): The Slowdown in European Productivity Growth: A Tale of Tigers, Thories and Textbook Labour economics, Northwestern University, Mixco.

Echevarria, C. (1997): Changes in Sectoral Composition Associated with Economic Growth”, International Economic Review, 38, 431-52.

European Commission (2003): Drivers of Productivity Growth: An Economy- Wide and Industry Level Perspective, EU Economy Review, Brussels.

EEA 2007, Greenhouse Gas Emission Trends and Projections in Europe 2007.

EEA 2008, EN21 Final Energy Consumption Intensity, European Environmental Agency EEA Report No 6/2008.

Evangelista, R. (2000): Sectoral Patterns of Technological Change in Services, Economics of Innovation and New Technologies, 9, 183-221.

Faggian A. and McCann P. (2006): Human Capital and Regional Knowledge Assets: a Simultaneous Equation Model, Oxford Economic Papers, 58, 475- 500.

Faggian A. and McCann P. (2009): Universities, Agglomerations and Graduate Human Capital Mobility, Journal of Economic and Social Geography (TESG), 100, 210-223.

Fatás, A. (1997): Countries or regions? Lessons from the EMS experience, European Economic Review, 41, 743-751.

Fernando and Yvonn.Q. (2008): Long-Run Economic Growth: Source and Policies, in (eds.) R. Glenn Hubbard, Anthony Patrick O`Brien, 2^nd ed.

Prentice Hall Business Publishing Economics.

Fisher, A. G. B. (1935): The Clash of Progress and Security. New York: Augustus M.

Kelle.

Friedman, M. and S. Kuznets (1945): Income from Independent Professional Practice.

New York: NBER.

Fuchs, V. (1980): Economic Growth and the Rise of Service ployment. Prepared for conference towards Explaining Economic Growth”, Institute of World Economics, Kiel.

Gary S. Becker (1964): Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education. Chicago, University of Chicago Press.

ISBN 978-0-226-04120-9.

Goldberger, A.S. (1972): Structural Equation Methods in the Social Science, Econometrica , 40, 979-1001.

Guerrieri, P., and Meliciani, V. (2005): Technology and International Competitiveness: The Interdependence Between Manufacturing and Producer Services. Structural Change and Economic Dynamics, 16, 489-502.

Görzig, B., Piekkola, H. and Riley R. (2010): Production of intangible investment and growth: Methodology in INNODRIVE.

Hartman, A., Farrello, D., Graubner, M. and Remes. J. (2008): Capturing the European Enrgy Productivity Opportunity. McKinsey Global Institute.

Hartwing, J. (2008): Productivity Growth in Services Industries: Are the Transatlantic Differences Measurement-Driven?, Review of Income and Wealth, 54, 494-505.

Ilmakunnas, P. and Piekkola, H. (2010): Production of Intangibles Investment and Growth Methodology in INNODRIVE (eds.)

Jimeno Moral, J. E. and Saiz, L. (2006): Structural Breaks in Labour Productivity Growth: The United States vs The European Union, Banco de Espana Working Paper, No. 0625.

Jorgenson, Dale. (1988): Productivity and Postwar US Economic Growth, Journal of Economic Perspectives, 2, 23-41.

Jorgenson, Dale. (1990): Productivity and Economic Growth. Harward University.Kaldor. N. (1966). Causes of the Slow Rate of Growth in UK.

Cambridge UP.

Romer, P. M. (1986): Increasing returns and long run growth, Journal of Political Economy, 94, 1002-1037.

Kongsamut, P., Rebelo, S. and Xie, D. (2001): Beyond BalanceGrowth, Review of Economic Studies, 68, 869-882.

Kratena, K. (2005): Sectoral Economy: Do Sectors Matter?, Estudios De Economia Aplicala, 23, 289-289.

Kuznets, S. (1957): Quantitative Aspects of the Economic Growth of Nations.

Industrial Distribution of National Product and Labor Forces, Economic Development and Cultural Change, 4.

Landes, David (1969): The Unbound Prometheuse. Cambridge, MA: Harward University Press.