An institutional approach and input-output analysis for explaining the transformation of the Turkish economy

Ünal, Emre Article An institutional approach and input-output analysis for explaining the transformation of the Turkish economy Journal of Economic Structures Provided in Cooperation with: Pan-Pacific Association of Input-Output Studies (PAPAIOS) Suggested Citation: Ünal, Emre (2018) : An institutional approach and input-output analysis for explaining the transformation of the Turkish economy, Journal of Economic Structures, ISSN 2193-2409, Springer, Heidelberg, Vol. 7, Iss. 3, pp. 1-38, https://doi.org/10.1186/s40008-017-0101-z This Version is available at: https://hdl.handle.net/10419/194908 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ An institutional approach and input– output analysis for explaining the transformation of the Turkish economy Emre Ünal* 1 Introduction In this paper, the main questions are: First, what were Turkey’s transformation periods, and what conditions furnished basis of technological changes? Second, what is Turkey’s growth model, and why it is still far from institutional complementarity? Two main concepts, growth models and institutional factors, can be used to define the economic and social changes in an economy. These concepts not only point out economic and political problems, but also effect a compromise between institutional changes and growth models, which serve to describe necessary arrangements inside an economy and help realize institutional complementarity between macroeconomic and institutional factors. Hence, using an institutional approach will enable exploration of the connection between regulation in a capitalist economy and its historical evolution Abstract The increasing inconsistency in the Turkish economy gave rise to transformations necessitated by the manner of its growth and technological changes. Between 1923 and 1962, the Turkish economy was based on agricultural growth strategies, which spurred development in the manufacturing sector and caused technological change. Between 1962 and 1985, the Turkish economy developed under import substitution industrialization, with regulations based on domestic consumption growth strategies. In the early 1980s, most of the developed countries, in parallel with which Turkey had development policies, implemented strong institutional changes via deregulation policies, but Turkey could not produce the necessary configurations in accord with open and export growth economies. The period between 1985 and 2003 was a long transformational period for Turkey vying to be a stable export growth country. This slow transformation was caused by path dependency and had been based on regulations that destabilized Turkey and deepened problems remaining from previous periods. Post-2003, the Turkish economy was able to develop more consistent export growth by forced institutional changes, but its macroeconomic factors still do not show institutional complementarity. Keywords: Agricultural growth, Domestic consumption growth, Export growth, Institutional factors, Stylized facts of structural transformation, Input–output analysis JEL Classification: C6, E6, F4, O4 Open Access © The Author(s) 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. RESEARCH Ünal Economic Structures (2018) 7:3 https://doi.org/10.1186/s40008-017-0101-z *Correspondence: [email protected] Department of International Trade, Faculty of Economics, Administrative and Social Sciences, Istanbul Gelisim University, Cihangir Mahallesi Şehit Piyade Onbaşı Murat Şengöz Sokak No:8, 34310 Avcılar/Istanbul, Turkey Page 2 of 38 Ünal Economic Structures (2018) 7:3 (Jessop and Sum 2006a, b: 1–3). Regulations emerge as a result of a change in a production system based on macroeconomic factors, which can be defined by growth models. As the production system changes in an economy, regulations usually remain slow, which causes an economy to follow radical institutional changes to create a compromise between the growth model and institutional factors. However, that does not mean each institutional change can create institutional complementarity based on macroeconomic factors. One of the reasons for the slowing of growth in the Turkish economy pre-2003 was its slow regulations under the dysfunctional governments. Radical regulations were practically established or supported after low economic performance in Turkey in the 1960s, which transformed the economy into one which was closed and protected, with import substitution industrialization based on domestic consumption growth after technological change from agricultural growth. In the early 1980s, increasing economic and political problems weakened government and gave way to new institutional changes and supported regulations for export growth. Nevertheless, the period of 1985–2003 remained a long transformation process for the Turkish economy. Finally, in the 2000s, regulations which made the Turkish economy stronger compared to that of the pre-2003 period were created by the government and supported by international organizations such as the IMF, the World Bank and the EU.1 The main reasons for economic crises in Turkey are flaws in its institutional factors, which conflicted with its macroeconomic factors. To avoid similar crises in future, institutional factors should be evaluated and reformed. Possible reforms that may occur in institutional factors could eliminate the influence of a crisis and stabilize the economy. Therefore, to create new institutional changes, the current growth model in an economy must be properly known and understood. In this work, Turkey and its growth strategies particularly from 1950 are considered. The Turkish economy went through several important periods. These periods were 1923–1962, 1962–1985, 1985–2003 and post- 2003. The main institutional changes that influenced and contributed to the future of the Turkish economy emerged in the periods after economic crises. The institutional changes used in this paper are inspired by the régulation theory. According to the régulation theory, there are five institutional factors: bank and credit relations, wage and labor relations, mode of competition, mode of international insertion and the role of government (Boyer 1990: 38–39; Boyer and Hollingsworth 1997: 49–54; Boyer and Saillard 2002: 44). However, these concepts encapsulate many factors. To make it more descriptive, the important concepts that influenced the Turkish economy over time are used, and summarized in Table8. The wage-labor relation is one of the important concepts of the régulation theory (Boyer and Yamada 2000: 10). The wage-labor relation can be influenced by income policies related to trade unions and modes of employment. Therefore, wage-labor relations consist of productivity growth, wage growth, and their interaction with inflation in the economy. The position of labor in sectors will be taken into account. Another important concept for the Turkish economy is related to exchange rate systems. The position of a country in international insertion can consist of exchange rate and competitiveness, which affect the trade balance, 1 The Justice and Development Party, Erdoğan’s government began implementing regulations as only governing party in the parliment. Page 3 of 38 Ünal Economic Structures (2018) 7:3 and is influenced by productivity growth and wage rate growth inside the economy. The production base of the important industries will be considered, and the role of privatizations in the economy will be highlighted. The mode of economy is depended on the macroeconomic perspectives of an open or closed economy. The government, with its regulatory intervention and role in the economy, will be considered. There are plenty of works in which to find institutional changes in the developed countries, and some works about the Latin American countries, which define their growth regimes and institutional factors (Aboites etal. 2002: 280–287). There have been some works which have analyzed the Turkish economy from an institutional perspective in order to explain its transformation in terms of macroeconomic and institutional factors, including wage-labor relations and exchange rate systems (Ünal 2016a, b, 2017). In that work, which differs from previous research, the Turkish economy is analyzed using broad concepts of institutional factors, and with a deeper theoretical and historical perspective, covering important periods of the economy when chronic problems emerged. In Sect.2, stylized facts of structural transformation are discussed. The macroeconomic growth models designed by input–output analysis and assumptions are explained. In Sect.3, major developed countries (Canada, France, Japan, the Netherlands, the United Kingdom and the United States) were briefly examined in terms of stylized facts of structural transformation, macroeconomic and institutional factors. In Sect.4, Turkey’s transformation from agricultural growth to domestic consumption growth is identified. Institutional factors in the period of agricultural growth are examined, and the reasons for the economic crisis in that period are discussed. Because of its parallel institutional factors, Turkey is compared with the developed countries by deriving input–output tables. Moreover, institutional changes that shaped domestic consumption growth in the period of 1962–1985 are explained, and the reasons for economic crisis under the domestic growth policies are clarified. In addition, the transformation into export growth is depicted in terms of institutional factors, and the importance of the 2000– 2001 economic crisis is extensively discussed. Finally, in Table8, macroeconomic and institutional factors, in the context of their periods, are clearly laid out. 2 Stylized facts of structural transformation and macroeconomic growth models Kuznets (1973) asserted that there are fundamental factors of economic growth that distribute economic goods to its population. These are advancing technology, and institutional and ideological adjustments. These factors change over time and regions. He highlighted facts about modern economic growth that cover rapid growth of population and per capita product, productivity growth, consumption and distribution of workers in sectors for structural transformation that is defined by a shift from the agricultural to the non-agricultural sector, and from the manufacturing to the service sector, and he considered the manufacturing sector a driving force for economic development.2 Kaldor (1961) highlighted stylized facts as regards the process of economic change and 2 Labor shift between sectors are important components of economic development. The countries which do not experience structural transformation between sectors generally remain poor and less developed. For sectoral classification, see Kuznets (1957). Page 4 of 38 Ünal Economic Structures (2018) 7:3 development in advanced capitalist countries. Some of these facts are about the distinction features of labor productivity, capital and output in economy. He considered exportled growth as a key factor for economic growth and pointed that the British economy lagged behind because of its consumption-led growth.3 For Solow (1970), technological upgrading is the source of increased productivity. In other words, productivity growth is an important measurement for describing structural transformation by defining the relation between output and input variables such as labor and capital. Verdoorn (2003) pointed out that productivity growth is connected with output and employment, and considered the productivity growth of the manufacturing sector as a significant factor for economic growth, and its elasticity with production in various countries.4 These works give clues to help us define structural transformation in economies. The standard transformation for an economy is from the agriculture to the manufacturing and service sectors, which result in increasing GDP, GDP per capita, productivity and a shift from low- and medium-tech industries to medium- and high-tech industries. In such a transformation, a developed industry leads labor to move to other developing industries. For instance, experiencing high productivity growth through low-cost production and mechanization in the agricultural sector enhances the manufacturing sector where labor desires to move and reproduce itself. This transformation creates a new dominant sector and contributes to economic development. In other words, structural transformation from the agricultural sector to the manufacturing sector gives rise to industrialization and develops a path toward a technological frontier. Economic development creates two economic conditions: “standard” and “revolutionary.” A standard economic condition indicates a transformation as workers move from agriculture to a developing manufacturing sector.5 Revolutionary economic conditions emphasize technological change and increasing productivity when workers become more sophisticated in the use of technologies and gain the organizational skills to engage in the domestic and export goods sectors, which contribute to economic development. In this respect, structural transformation also means more complicated production processes and higher productivity growth. Structural transformation creates new leading sectors, leads to new organization of workers and facilitates technological upgrading. What define a standard structural transformation are the stylized facts: Sectoral shares of workers, value-added, production, and export. 2.1 Describing structural transformation with macroeconomic and institutional factors From the 1950s to the early 1970s, national data, which gave detailed information about production in the agricultural, manufacturing and service sectors, and employment 3 For additional information see King (2009: 77) and Uni (2007). 4 There are also some other works considered modern economic theories to explain structural transformation by using stylized facts. Herrendorf etal. (2013) used employment and value-added shares in sectors and pointed the developed countries usually experiences downward process in the value-added share of manufacturing after it reaches a certain extent, whereas agriculture generally decreases. Thus, the service sector gains importance. Michaels etal. (2012) examined structural transformation by six stylized facts to describe population shift across rural and urban areas. Briones and Felipe (2013) briefly mentioned that although developing Asian countries experienced structural transformation, agriculture has kept its important position. The sector’s employment share and output were still at significant level. 5 This technological transformation can also be explained by flying geese theory. Turkey experienced large amount of foreign direct investment inflows from developed EU countries such as Germany and Netherlands in the 2000s. For additional information, see Ünal (2017). Page 5 of 38 Ünal Economic Structures (2018) 7:3 statistics, was used to define technological change and structural transformation in Turkey. The input–output tables of Turkey were derived from TurkStat and WIOD. The input–output tables used from TurkStat cover the years 1973 and 1985. Whereas the 1973 table is a descriptive source for indicating domestic consumption growth, the 1985 table is a source of describing technological change in comparison with the previous and subsequent period. TurkStat released detailed (64 ×64) tables but did not provide employment information. Hence, the tables were aggregated to nine main industries, and TurkStat employment data was used to calculate the productivity growth of nontradable and export goods sectors.6 Although WIOD 2003 and 2011 (35×35) tables provide employment data, the sectoral price deflators were not provided for non-trada- ble and export goods; thus, the deflators were collected from UN data. For this kind of study, although there was some lack of data, additional sources helped the study became more substantial. Had TurkStat included detailed employment information, it could have helped calculate productivity growth; and had WIOD released sectoral price deflators for export industries it would even have been possible to analyze the role of industries in the export goods sector. Nevertheless, the tables are sufficient to show technological change and the shift in dominance between the non-tradable goods and export goods sectors. Although it is controversial to compare a developing country such as Turkey with developed countries, it is a fact that Turkey experienced macroeconomic factors and followed institutional changes in parallel with developed countries, whether successfully or not. Moreover, OECD input–output tables for developed countries provide an opportunity to compare Turkey with these countries.7 The Turkish economic transformation can be analyzed by commodity base productivity growth calculation via input–output tables. This method of calculation gives information about dynamic developments in the economy and helps understand the role of the productivity growth of export goods, which is the driving force of economic growth in export growth countries. Furthermore, increasing productivity in the export goods sector means that much more complicated products begin to be transacted, namely, technological change emerges related to a revolutionary economic condition. To find the productivity growth rates, labor input is used to divide sectors into non-tradable goods and export goods, which are specific conceptions intended to create macroeconomic factors to explain structural transformation from the 1970s onward. Although technological transformation plays a crucial role in enabling developing economies to catch up, lack of proper institutional changes can be a barrier. In addition, different development paths require different institutional forms. Therefore, unfavorable institutional changes can cause slow economic development and slow structural transformation. Conclusively, there is a standard development path for developing countries, but if institutional factors are not defined properly, this causes slow transformation and the country lags behind. In other words, the development potential of countries might be limited. To understand the Turkish economic transformation better, macroeconomic 6 For employment data see TurkStat, statistics of employed persons by type of economic activity, report “Statistical Indicators 1923–2011.” 7 It is difficult to find 1970s’ input–output tables for developing countries because most of them were the part of communist economy. Page 6 of 38 Ünal Economic Structures (2018) 7:3 growth models are considered in order to identify institutional problems. However, eliminating potential problems requires a macroeconomic growth model to define economic conditions. These models are mostly categorized by macroeconomic factors to provide a favorable analysis, so the institutional factors suitable for a structural transformation must be well-understood. So the fundamental structural transformation paths categorized in this paper are agricultural growth, domestic consumption growth and export growth. 2.2 Macroeconomic growth models Agricultural growth emerges at the first stage of industrialization, usually spurring the manufacturing sector to grow and become the leading sector in an economy in subsequent years. One of the defining features of an agricultural growth country is that its productivity growth is greater than that of manufacturing. After experiencing a technological change to the manufacturing sector, complicated commodities are used in the production process; thus, labor cost, import material costs and mark-up rates gain prominence. The growth models assume that there is proportional wage growth between the nontradable and export goods sectors. According to Table9, all seven countries had proportional wage growth between the two sectors. Moreover, the assumptions of a constant mark-up rate and negligible import costs are important variables in the models. However, differences between the price levels emerge due to proportional wage growth,8 but disproportional productivity growth of non-tradable goods and export goods.9 To explain more clearly, two countries are used in the models. Country A is shown with superscript A and country B with B. The one of the reasons that Turkey had a closed economy in the 1970s was because of a global trend in economic policies. Developed countries mostly had high proportional wage rate growth that surpassed their productivity growth rates, and experienced high inflation. In other words, they had a “domestic consumption growth model” which means the productivity growth of non-tradable goods (ˆ qA n ) was higher than that of exports goods (ˆ qA e ). Subscript n and e indicates non-tradable goods and export goods, respectively. (^) indicates growth rate (see appendix in detail). The economic model of the 1970s can be described as follows: 8 The price level (p) is based on vertically integrated labor input coefficients (v) (see Appendix for the calculation method), nominal wage rate (w), mark-up rate (m), and imported material cost (cim) as follows: p=(1+m) (wv+cim). Here, price is divided into the price of non-tradable goods and export goods. Subscript n and e represents non-tradable and export goods respectively as follows: pn=(1+mn) (wvn+cim) and pe=(1+me) (wve+cim). To get the purchasing power parity (PPP), the export price is considered between two countries (A and B) as follows: PPP×export price of country A=export price of country B, q indicates productivity and (ˆ) indicates growth. As the part of assumptions, in conditions of negligible import cost, PPP was calculated between the two countries as follows: P ˆ PPA=   1+mB e  +ˆ wB−ˆ qB e  −   1+mA e  +ˆ wA−ˆ qA e  . PPP is the hypothetical exchange rate. Price levels can be calculated from the United Nations data “National Accounts Estimates of Main Aggregates” (http://data.un.org/ Explorer.aspx?d=SNA). For export price level, export at current prices (national currency) and export at constant prices (national currency) are considered. The price level of non-tradable goods was calculated by considering respectively both current and constant prices; domestic demand=GDP−export+import. 9 Proportional wage rate growth can be approximately calculated by considering change rates in price levels and productivity growth, calculated via input–output analysis as follows: ˆ w e=ˆ pe+ˆ qe−   1+me  and ˆ w n=ˆ pn+ˆ qn−   1+mn  . (1) ˆ wA >ˆ qA n >ˆ qA e and ˆ qB e =ˆ qB n =ˆ wB Page 7 of 38 Ünal Economic Structures (2018) 7:3 In model (1), in country A, because high social welfare systems served to protect trade unions and support domestic consumption, the proportional wage rate growth ( ˆ wA ) is larger than the productivity growth of non-tradable goods. Also, the productivity growth of non-tradable goods is greater than the productivity growth of export goods because international trade is secondary, and not promoted, which decreases the efficiency of export goods sector, causing slow productivity growth. In the country A, inflation is high, and it is less competitive under the fixed exchange rate system compared with that of country B. Country B shows a perfect economy under equal macroeconomic factors. Export growth became dominant in the 1980s, when many countries abandoned closed economic policies and focused on export growth to increase their competitiveness in international trade. To do that, countries scrapped welfare programs and reduced the power of trade unions to eliminate inflation and decrease unit labor cost (ULC).10 Therefore, in the 1980s, wage rate growth showed a declining trend compared with the 1960s and 1970s. Export growth models can be explained by a low-cost production model and the Balassa–Samuelson model. The Eq.(2) shows a “low-cost production model.” In that model, wage rate growth is fixed to the productivity growth of non-tradable goods, which creates low ULC growth in the export goods sector and hence increases competitiveness in the international context. Under a floating exchange rate, country A has increasing PPP, which means the country will experience appreciations in its national currency because the change in the exchange rate moves in parallel with that of PPP. The Eq.(3) reflects the “Balassa–Samuelson model.”11 The assumption of this model is based on proportional wage growth, but disproportional productivity growth. In that model, wage rate growth is fixed to the productivity growth of export goods and the productivity growth of export goods is larger than the productivity growth of non-tradable goods under the fixed exchange rate system. In that model, because of disproportional productivity growth (export-biased productivity growth) between two sectors, high inflation emerges.12 2.2.1 The over‑valued currency model and institutional complementarity When the other countries in international trade shifted to the low-cost production model and the Balassa–Samuelson model in the 1980s, Turkey, which could not implement necessary institutional changes, fell into the “over-valued currency model,” with high productivity growth of export goods compared with that of non-tradable goods. If the country cannot implement necessary institutional changes, because of highcost production and an over-valued currency, it will produce high inflation, lose its 10 The method of ULC growth rates of export goods and non-tradable goods: Uˆ LCe =ˆ we −ˆ qe and Uˆ LCn =ˆ wn −ˆ qn . (2) ˆ qA e >ˆq A n =ˆ w A and ˆq B e =ˆq B n =ˆ w B (3) ˆ wA =ˆq A e >ˆq A n and ˆq B e =ˆq B n =ˆ w B 11 For additional information see Balassa (1964), Samuelson (1964), Uni (2012) and Ünal (2016b). 12 For additional information about EMU and export-led growth models based on analysis in the 2000s, see Ünal (2016a, b). Page 8 of 38 Ünal Economic Structures (2018) 7:3 competitiveness, have a high trade deficit and create an economy, which is exposed to economic turmoil and large devaluations in its exchange rate. The over-valued currency model is defined as follows: In that model, the exchange rate system can be fixed or floating, but is repressed by central banks to remain over-valued. That model basically reflects the Turkish economy in the post-1980s. An economy which has that growth model suffers because of high inflation, dramatic depreciations in the exchange rate, low competitiveness and a high trade deficit relative to country B. In that model, a national currency is over-valued, which means its appreciation rate is higher than that of PPP. To generate institutional complementarity between macroeconomic and institutional factors for export growth, wage rate growth must fall to at least that of productivity growth of export goods. 3 Structural transformation in major developed countries Figure1 shows the relationship between the distribution of workers among the three main sectors and their movements with GDP in the major developed countries. The countries were chosen for the analysis because of their available input–output tables in the same periods to discuss development process in the 1970s and 1980s (see Table1). What is generally seen on Fig.1 is consistent decline in the sectoral share of workers in the agricultural sector, whereas steady increase in that of workers in service sector. Furthermore, the sectoral share of workers in the manufacturing sector in the developed countries except in Japan indicates more visible downward inclination toward the 1970s. The sectoral share of value-added in the three sectors in all countries moved differently (see Fig.6 in Appendix). Whereas the sectoral share of manufacturing sector indicated downward process in the 1970s that of service sector was seen as attracting more workers and contributing more for value-added. In addition, the sectoral share of production in the manufacturing sector decreased and that in service sector became dominant toward the 1990s in the developed countries (see Table10 in Appendix). As these countries had already developed manufacturing sector, the sectoral share of export in the manufacturing sector compared with others were more significant.13 The curves of the countries discussed in Fig.1 showed similar patterns in the selected years. Furthermore, the macroeconomic factors of the developed countries moved almost in parallel with each other in the 1970s and 1980s. In these countries, wage growth, which had been supported by welfare state policy, was high. In this kind of policy, the main aim was to give workers purchasing power to provide effective demand for domestic production, and this influenced the productivity growth of non-tradable goods. Thus, inflation and labor cost growth were high. Productivity growth was based on the non-tradable goods sector because domestic consumption was the main policy (4) ˆ wA >ˆq A e >ˆq A n and ˆq B e =ˆq B n =ˆ w B 13 The contribution of the service sector to export was very low in Turkey compared with those of developed countries. For instance, the sectoral share of exports in the service sector was 17.0 in Canada, 23.3 in France, 17.3 in the United Kingdom, 15.8 in Japan, 26.4 in the United States in 1990, and 19.9 in Netherlands in 1986, but in Turkey it was just 0.7 in 2014. As the major countries had developed industrialization long before these periods, there was no clear indication that could help define structural transformation by the ratio of export to import, but sectoral wage growth was usually higher in the service sector than in the manufacturing sector, and that of the manufacturing sector was higher than that of the agricultural sector. Source: OECD input–output tables. Page 15 of 38 Ünal Economic Structures (2018) 7:3 had an agricultural growth regime from 1923 to 1962, as seen on Table4, where productivity growth in the agriculture sector was greater than those of other sectors. The turning point for industrialization that supported the manufacturing sector emerged in the 1950s. After the Democrat Party was elected in the 1950s, it followed liberal policies in the Turkish economy, and thus opened the economy for international trade and supported private industries in developing its agricultural sector. Hence, the agricultural production became dominant growth strategy as a flash point for mechanization and industrialization in the economic development of Turkey in the 1950s. Table4 shows that the productivity growth of the agricultural sector was larger than that of the manufacturing sector, and inflation was very low between 1950 and 1956. That indicates the economy was growing through agricultural production. The productivity growth of the agricultural sector was 4.3%, whereas that of the manufacturing sector was 2.2%. Inflation stood at 4.2%—one of the lowest rates in Turkish economic history. In that period, agricultural production was dominant and it was the main contributor to economic growth. The contribution of agricultural production to all sector’s productivity growth can be seen clearly: 4.1%, closer to that of agriculture. In the period 1956–1962, Turkey experienced an economic crisis which decreased its performance in terms of productivity growth. The productivity growth of the agricultural sector was 2.4%, and that of the manufacturing sector, 1.8%. Productivity growth of all sectors remained at 2.9%, while productivity growth of the service sector was just 0.3%. In that period, falling productivity growth and an economic crisis boosted inflation, which increased to 11.8%. Low export growth caused a trade deficit and deepened Turkey’s problems into a full-blown economic crisis in the mid-1950s. For instance, in 1946, the ratio of export to import was 1.8, but this ratio decreased to 0.63 in 1955. Between 1950 and 1953, imports increased by approximately 18.3% and exports by 12.6%. The peak years were 1951 and 1952, when imports rose by 40.8 and 38.3%, respectively.19 19 Source: Turkstat (foreign trade by years). Table 4 Productivity growth in agricultural, manufacturing, service, and all sectors (annual rate, unit: %). Source: Author’s calculation. Employment data was derived from TurkStat (employed persons by type of economic activity). Productivity growth was calculated by deriving sectoral data from the Ministry of Development’s report (Economic and Social Indicators between 1923 and 2014) and calculated with real outputs. Inflation was derived from TurkStat (consumer price). The inflation data between 1923 and 1939 replaced with wholesale price index The italic emphasis the period when the manufacturing sector became the dynamic sector Period Agriculture Manufacturing Service All sectors Inflation 1923–1931 8.3 3.9 3.8 6.4 − 3.3 1931–1939 6.2 0.7 4.5 5.0 2.3 1939–1945 − 9.4 − 8.8 − 6.5 − 7.7 19.7 1945–1950 4.2 1.9 1.1 4.6 0.3 1950–1956 4.3 2.2 − 4.6 4.1 4.2 1956–1962 2.4 1.8 0.3 2.9 11.8 1962–1973 1.4 3.8 1.8 3.9 8.2 Page 16 of 38 Ünal Economic Structures (2018) 7:3 4.1.1 Institutional factors in the period of agricultural growth One of the reasons for the high productivity of the agricultural sector compared with other sectors between 1950 and 1956 was government policies to invest in agriculture and produce for global demand, which brought technological changes to the Turkish economy, when the first imports of high-tech transportation and machines were demanded, and helped develop mass production in subsequent years. This transformation was simultaneously followed by institutional changes based on liberal policies aimed at creating a non-restricted economy shaped by state-led policies to ease the import and export of goods in international trade. Increasing imports in the Turkish economy necessarily gave way to the needs for exports as well. Therefore, as a country, which did not have enough industrial development, agriculture gained prominence in economic activities. The increasing amount of machinery lessened demand for labor in the agricultural sector, but created a need for physical power in manufactures. Hence, labor was attracted by developing manufactures through migration from rural areas. For the mode of employment, as seen on Fig.2, the sectoral share of workers in the manufacturing sector increased significantly. For instance, whereas the number of workers in the agricultural sector increased on average by 0.9%, it was 5.6% in the manufacturing sector between 1950 and 1962. However, during that period, the number of workers in manufacturing constituted a small proportion of the total employed labor force, which shows that the mode of employment was based on the agricultural sector. Whereas the percentage of agricultural workers was 84.3%, that of manufacturing workers was only 8.6% of the total number of employed people in 1950. Although the number of agricultural workers increased consistently until 1990, as a percentage of employed people, it decreased. In this period, Turkey did not have institutionalized trade unions. The bulk of labor consisted of agricultural workers during the 1950s, and economic growth was based on agricultural production between 1950 and 1962. Turkey was able to develop and establish private industries and increase production by agricultural land development. The lack of trade unions in the Turkish economy did not influence inflation because wage rate growth did not translate into pressure for production and government policies. Inflation was very low between 1923 and 1956, except during the Second World War, which could be explained by high productivity growth in the agricultural sector and low labor pressure for wage increases.20 The relatively lower cost production compared with following years, and open economic policies, supported demand for Turkish agricultural products in international trade. However, between 1956 and 1962, productivity growth decreased; thus, inflation climbed significantly, to 11.8% (Table4).21 In that period of 1950s, there was a fixed exchange rate system based on the US dollar. The Government exerted excessive influence on the CBRT to restrict fluctuations in the exchange rate. The inflation rate, which has been one of basic issues in the Turkish economy, did not cause great problems in the early 1950s. 20 Although Turkey did not enter the Second World War, it was affected negatively. Productivity growth rates decreased significantly and inflation rose dramatically between 1939 and 1945 (see Table4). 21 It is generally accepted that Korean War between 1950 and 1953 was one of the reasons that increased demand for agricultural products of Turkey. Page 17 of 38 Ünal Economic Structures (2018) 7:3 Although the economy was shaped by liberal economic policies, with non-unionized agricultural and manufacturing workers, who did not have a basic right to strike, wage growth policy was implemented by government intervention. Nevertheless, decreasing productivity growth and a high trade deficit problem led to significant devaluations in the lira, as seen in Fig.3. In the late 1950s, the lira depreciated by approximately 73.9%, and in 1960, it depreciated by approximately 84.8%. Although, owing to non-institution- alized trade unions, wage rate growth did not constitute serious pressure on inflation, the increasing trade deficit, the long period of fixed exchange rates, and compounding economic problems related to high costs of production and inflation gave way to devaluations. The base of production was developed around agricultural production and that shaped Turkey’s configuration in international trade. Turkey could not establish its manufacturing industries sufficiently to produce durable goods for the needs of the middle class or to export. However, agricultural base production gave way to increasing development in manufacturing industries over the years. In that period, Turkey had open economic policies and for the first time experienced a high trade deficit, which became a chronic problem in subsequent periods (see Fig.4). Increasing international demand for agricultural products became a reason to follow open import and export policies. The government’s open economic policies involved supporting the growth of private industries alongside public industries. Thus, the role of private industries in production increased significantly. As seen in Table5 the open economic policies and the effect of promoting private industries is clear. In the period of 1950–1960, the number of private -1.8 -1.3 -0.8 -0.3 0.2 1923 1926 1929 1932 1935 1938 1941 1944 1947 1950 1953 1956 1959 1962 1965 1968 1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010 2013 Fig. 3 Change rate in the lira (annual rate, 1923–2013). Source: Author’s calculation. The change rates in the lira were derived from TurkStat (Statistical indicators 1923–2013) 0 50 100 150 200 1923 1926 1929 1932 1935 1938 1941 1944 1947 1950 1953 1956 1959 1962 1965 1968 1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010 2013 Fig. 4 Ratio of export to import in the Turkish economy (1923–2013). Source: TurkStat (foreign trade by years) Page 18 of 38 Ünal Economic Structures (2018) 7:3 manufactures increased by approximately 9.4%, higher than that of public manufactures, which increased by 7.6%. In the early 1950s, the number of industries increased significantly, not only because of demand for agricultural products, but also because of the mode of economy was based on open and liberal policies. Although in following years deepening economic instability slowed growth in the number of industries, the Turkish economy was able in relative terms to develop its industrial production base to follow import substitution industrialization policies in an attempt to reduce economic problems which remained from the period of agricultural growth.22 In other words, the failed attempt at an open economy provided an opportunity for industrialization. Although private industries became an important part of the economy, privatizations were not visible and not yet institutionalized in the Turkish economy. The significance of private industries declined in the 1960s. Closed and protected economic policies in that period caused a slow increase in the number of private manufactures by 1.3%, whereas the number of public manufactures increased by 4.4%, (see Table5). In particular, in the early 1960s, the rate of change in the number of private manufactures decreased by 34.9% in Turkey. The role of government was shaped by open and liberal economic policies. The economy, which lacked trade unions and social movements, enabled the government to implement liberal policies with ease. Thus, government intervention in the market was distinctive and strong. The role of government was based on agricultural growth because government promotion of agricultural production that was the only economic means available in the country for the increase of export goods in the 1950s. Moreover, Turkey did not have a serious problem with inflation in the beginning of the 1950s, and was able to create high productivity growth, which made it relatively successful. Nevertheless, agricultural production became problematic in subsequent years due to low productivity growth caused by decreasing demand. This forced the country to implement devaluations in order to increase its competitiveness and cover its trade deficit. These problems could not be eliminated and continued in subsequent years. The government’s liberal policies were criticized by the military and its veterans after productivity growth decreased and the country encountered high inflation and a long-lasting trade deficit problem. The democratic government failed to produce effective policies to regulate the economy. Finally, the government lost its function in the end of 1950s, and, following the military coup, the prime minister was executed and a new government instituted regulation policies that turned Turkey into a closed and protected economy similar to those of the developed countries, promoting domestic consumption until the 1980s. 22 In the period of 1950–1956, the number of industries increased by 9.9%, and in the period of 1956–1962, increase was 0.3%. Source: Author’s calculation. Data was derived from TurkStat (manufacturing industry, total). Table 5 Change rate in number of private and public manufactures (annual rate, unit: %). Source: Author’s calculations. Manufacturing industries of public and private sectors were derived from TurkStat Category 1950–1960 1960–1970 Private manufactures 9.4 1.3 Public manufactures 7.6 4.4 Page 19 of 38 Ünal Economic Structures (2018) 7:3 4.1.2 The crisis of the agricultural growth regime In the beginning of the 1950s, Turkey was able to stimulate its agricultural productivity growth, which was the dynamic for low inflation, a stable exchange rate and industrial development. Between 1950 and 1953, the productivity growth of the agricultural sector was 10.2%, whereas that of the manufacturing sector was 3.2%. However, toward the end of the 1950s, Turkey encountered macroeconomic problems that deepened instability in the economy and led to essential reform measures taken in 1958. These problems occurred because of a high trade deficit and inflation following falling productivity growth. To cut demand for imports of goods and accelerate the export of goods, the lira depreciated significantly against the US dollar. Rising inflation became one of the reasons for the economic crisis at the end of the 1950s. To counter high inflation, an attempt was made to control monetary expansion by preventing the CBRT from financing public enterprises by printing money. These reforms failed to work as expected that followed with a structural change in the economy by leading to import substitution industrialization by 5-year development plans. In the beginning of the 1960s, following the military coup against the Democrat Party, new legislation was enacted. Turkey began implementing planning programs to improve its industrial production using import substitution industrialization strategies. Thus, agricultural production lost its importance and manufacturing production became the dominant base of the Turkish economy. In the period of 1962–1973, the productivity growth of the agricultural sector was 1.4% lower than that of the manufacturing sector, which was 3.8%. Furthermore, the productivity growth of the service sector was 1.8%, and all sectors’ productivity growth was 3.9% closer with that of the manufacturing sector. The increase in productivity growth in the manufacturing sector was larger than that of the agricultural sector highlights important technological changes in the Turkish economy, and a growing capacity in product manufacturing (see Table4). 4.2 Domestic consumption growth and institutional factors in the period 1962–1985 The main aim of import substitution industrialization is to achieve economic self-suf- ficiency by protecting infant industries until they reach a technological level whence they can compete in international trade. By this means, domestic industries grow and become more self-sufficient. This type of industrialization was implemented in Turkey using a series of 5-year development plans. The government implemented tariffs and quotas to protect domestic industries and give time and support for their development. In the 1960s and 1970s, governments stimulated investment to reach the objectives of their development plans. Some important industries such as the automobile, machinery, chemical and chemical products and electrical and optical equipment industries were established and upgraded in this period under the 5-year plans. Finally, the manufacturing sector became the base of Turkish exports from the 1980s onward. Compared with developed countries, Turkey had highest ULC growth in both nontradable goods and export goods. In line with other countries, because of closed economic policies, Turkey pursued domestic consumption growth. Between 1973 and 1985, wage rate growth was 34.2%, and the productivity growth of non-tradable goods was 3.8% higher than that of export goods, which was 2.0%. Moreover, Turkey’s inflation rate was greater than that of other countries. During the period 1985–2003, the most Page 20 of 38 Ünal Economic Structures (2018) 7:3 important point in Turkey contrary to other countries, wage rate growth increased significantly, to 66.2%, and inflation climbed to 63.5% (see Table1). Although Turkey was able to decrease wage rate growth to 11.2% between 2003 and 2011, it remained higher than the productivity growth of export goods and non-tradable goods. Hence, Turkey came under the dominance of the over-valued currency model due to its over-valued currency and high-cost production. An analysis of Turkey’s macroeconomic factors showed that Turkey had parallel macroeconomic factors with the developed countries in the 1970s as a protected economy. In the 1980s, in contrast to the developed countries, its wage rate growth was stimulated and its inflation boosted. However, Turkey became an export growth country after 1985, which shows that there were contradictions in its macroeconomic and institutional factors, which were not designed according to economic policies in these periods. Turkey could not create institutional complementarity to satisfy export growth under free and open economic policies designed to increase its competitiveness. Turkey should have at least decreased its wage rate growth to match the productivity growth of export goods for institutional complementarity that did not break international competition and could have decreased ULC growth in Turkey. The domestic consumption growth was supported through forcible regulations in Turkey. To conduct the first 5-year development plans for developing basic and intermediate goods industries in order to reach the desired level of industrial developments in specific, planned years, a state planning organization was established in September, 1960 to assist the government. The institutional factors of the 1950s evolved alongside increasing manufacturing production, and an international context based on the idea of the developed capitalist economies employing government intervention in the economy to control prices and incomes, and to support the rights of workers and trade unions through welfare policies. Nevertheless, unfavorable macroeconomic factors prevented Turkey having a stable growth model. Turkey had similar problems to the developed countries, such as inflation and high ULC. 4.2.1 Institutional factors in the period of domestic consumption growth During the period 1962–1985, wage-labor relations changed significantly and new modes of employment emerged. Although agriculture kept the bulk of the labor force, the importance of manufacturing workers increased for import substitution industrialization. Thus, trade unions came into prominence, and gained legal ground owing to growing manufacturing production and demand for workers. The government postulated the idea that manufacturing production was more important than agricultural production as manufactures contributed more to the economy in terms of export than agriculture could. This prompted an increase in the number of manufacturing industries, which attracted workers in the cities and agricultural workers from rural areas. Nevertheless, agriculture remained an important part of the economy. The mode of employment was thus based on both agricultural and manufacturing workers. After the technological change from agricultural production to manufacturing production, Turkey created a legal status for trade unions. Growing industries furnished a need for the establishment of decent working conditions and grounds for social rights for workers; hence, collective bargaining became more of an issue in income policy. Page 21 of 38 Ünal Economic Structures (2018) 7:3 During the planning periods of the 1960s, collective bargaining became for the first time an aspect of manufacturing production, and legally recognized trade unions were able to conduct strikes and organize other industrial actions. After gaining basic rights, the trade unions began to flex their muscles in the early 1960s. Eight strikes occurred in 1963, and this number increased dramatically in following years (see Table6).23 Trade unions were influenced by radical left wing movements, in particular, the communism of the Soviet Union, and led to widespread political unrest in Turkey. Economic problems, combined with labor and political actions, devastated efficiency in production, with many workdays lost in industries in the 1970s. The increasing number of strikes and lock-outs inevitably damaged the Turkish economy and its industrial production. The pressure of the trade unions for wage increases led to high inflation, which was also compounded by the government’s monetary expansion to finance public industries by the CBRT printing money without counterpart. The government accepted minimum wage policies in the mid-1970s, based not on collective bargaining, but on workers’ own basic rights. In this period, more than 50% of workers were low-skilled, which means the minimum wage covered a large area in the economy. Hence, the institutional changes in the 1960s made trade unions and social movements stronger, but these reforms delayed new institutional changes in the 1970s and 1980s, namely the deregulation policies demanded by neoliberalism, and influenced the economic development and competitive position of Turkey in forthcoming periods. The exchange rate system was a fixed system. However, the lira experienced significant devaluations caused by the high ULC growth of export goods because it increased dramatically while productivity growth remained low. For instance, the ULC growth of export goods was 32.2% in the period 1973–1985 (see Table1). In that period, due to high production costs, the Turkish economy was unable to compete and eliminate its trade deficit problem. High wage rate growth over the productivity growth of nontradable goods caused destructive inflation, and high ULC growth of export goods became a major factor in deepening pressure on the lira and following its depreciations. In that period, the government intervened heavily in the policy of the CBRT over the exchange rate. Furthermore, high inflation, the high ULC growth of export goods and a 23 In Turkish, Grev ve Lokavt uygulamaları (http://www.csgb.gov.tr/csgbPortal/csgb.portal?page=grevlokavt) accessed on January 6, 2016. Table 6 Strikes and lock-outs in Turkey. Source: Data were derived from Ministry of Labour and Social Security (in Turkish, grev ve lokavt uygulamaları) Year Number of strikes Workdays lost to strikes Number of lockouts Workdays lost to lockouts 1963 8 19.739 0 0 1970 72 220.189 4 98.229 1975 116 668.797 7 67.949 1980 220 1.303.253 21 682.843 1987 307 1.961.940 221 484.572 1990 458 3.466.550 41 1.188.091 1995 120 4.838.241 5 162.512 2000 52 368.475 2 32.760 2012 8 36.073 0 0 Page 22 of 38 Ünal Economic Structures (2018) 7:3 compounding trade deficit strongly influenced the decisions of the CBRT, which engineered successive devaluations. Thus, the lira became under-valued between 1973 and 1985 in long-term (see Table2). The production base changed significantly as the economy evolved from the agricultural to the manufacturing sector. This technological change led to a large increase in productivity in the first years of the 1960s. For instance, between 1962 and 1964, the productivity growth of the manufacturing sector was 15.5%, whereas that of the agricultural sector was 6.1%, and inflation was 2.5%. However, unlike in previous periods, the manufacturing sector was protected owing to new institutional changes. A significant amount of industry remained in the public sector, under government control. Hence, the income policies of these industries were exposed to industrial action and the populist aims of the government. That means that the formulation of wage policies was based on individual interests rather than economic conditions. The institutional changes led to a protected and closed economy. Privatization was not institutionalized and was not a matter for political discussion. The influence of public industries around the country was distinctive. Import substitution industrialization was the dominant strategy, aimed at developing domestic industries via government policies. The mode of economy was shaped through protectionism, via taxes and tariffs and closed economic ideas based on domestic manufacture–a change from the open and liberal formation of the previous period to a protected and closed formation. This was an important complementarity which combined institutional factors; a potential basis for a consistent economy. For instance, protected industrial and closed economic policies as a mode of economy were important signs of institutional complementarity for the promotion of domestic consumption. The government’s intervention in the economy was excessive, and the role of government was important for promoting domestic manufacturing, but although the importance of manufacturing increased, international insertion was based on the export of both manufacturing and agricultural products in the 1960s and 1970s. Between 1960 and 1980, developing manufacturing production made Turkey more integrated into the global economy because of increasing interaction between countries based on imports and exports. However, in the 1970s, the ratio of exports to imports fell to its lowest point, approximately 30%. Figure4 shows this significant decrease in the ratio. In the 1950s, developments for industrialization under agricultural growth increased the import of goods, which, combined with the liberal policies of the government, resulted in a relatively open and unprotected economy. In the 1970s, a trade deficit was stimulated, inflation increased and Turkey experienced a new economic crisis. This problem was mainly caused in wage-labor relations related to income policies, which boosted inflation and ULC in the economy. Reforms to solve these economic problems could not be implemented efficiently and the economic decisions that had to be taken in the parliament remained slow. Necessary institutional changes did not emerge in time, and Turkey could not respond adequately to the global economic situation. The institutional changes that were developed in the 1960s and 1970s were not satisfactory for the Turkish economy because they brought excessive protection for employees, based as they were on welfare policies and shaped overwhelmingly by domestic consumption growth. These changes generated inconsistencies and caused deepening and chronic problems. Page 23 of 38 Ünal Economic Structures (2018) 7:3 Moreover, this situation prevented institutional complementarity in the macroeconomic base and caused path dependency in subsequent periods. 4.2.2 The crisis of domestic consumption growth regime Although the government supported domestic manufacture, due to the lack of complementarity between macroeconomic and institutional factors, industrial policies contributed nothing to the alleviation of problems in the Turkish economy, as had been expected in the 1960s, and this dragged the country into a devastating economic and political crisis at the end of 1970s. Turkey’s economic reforms were slowed down by political turmoil and social movements. Shaky coalition governments could not produce necessary economic reforms, and this deepened the country’s economic problems. Moreover, by decisions in January 24, 1980, the government imposed less restriction on imports of goods, supported export by promotions and implemented devaluations in the lira as a preparation for export growth strategies.24 The decisions taken in the 1980s were a result of a program supported by the IMF. The reforms were necessary to cut monetary expansion and aimed to reduce the power of trade unions, to increase privatization and decrease the influence of public industries in the market. Furthermore, Turkey decreased its support for the agricultural sector while promoting manufacturing production in an attempt to stimulate exports. This transformation, however, remained slow. In the beginning of 1980s, political and economic problems led to another military coup. The deregulation policies in the economy continued and these policies were relatively supported in following years. The trade union legislation that was established in the 1960s made it hard to achieve flexibility of labor without political unrest. In the 1980s, although trade unions lost a measure of their power, public industries still dominated the economy. Hence, the change to a free market economy remained largely incomplete. Additionally, expected privatizations of the largest public industries met with protests and other pressures, and could not be institutionalized until the mid-1990s when Turkey established an institution for the support of foreign direct investment into the public industries. Turkey began increasing its economic reforms to promote export growth in the economy. The export growth strategies began taking root after 1985 after a new democratic government took power. In consequence, the productivity growth of export goods increased more than that of non-tradable goods. Therefore, since the mid-1980s, export growth has become a distinctive feature of the Turkish economy. However, institutional changes remained uncompleted and did not show complementarity with macroeconomic factors in parallel with the developed countries. 4.3 Transformation to export growth in the period 1985–2003 During the period 1985–2003, Turkey was unable to implement strict institutional changes, and in 2000–2001 met with a severe economic crisis. For a long period of time, it suffered from high inflation, wage growth and unstable exchange rate. In other words, although Turkey gained a new production base for export growth from technological change, it was unable to develop new solutions, and this caused conflict between 24 In particular, after the collapse of the Bretton Woods System, free market economy and exchange rate systems became important components of institutional changes imposed by the IMF. Page 24 of 38 Ünal Economic Structures (2018) 7:3 macroeconomic and institutional factors. Hence, this period was basically one of transformation, when Turkey began seriously implementing deregulation policies. In the transformation period, Turkey failed to implement a number of institutional changes to prepare the economy for export growth, whereas the developed countries discussed in Table1 were able to do that by decreasing their ULC growth via deregulation policies. The wage rate growth was 66.2% greater than the productivity growth rates and inflation increased to 63.5%. The worst condition was that ULC growth of export goods in the Turkish economy increased from 32.2 to 60.9% between the periods of 1973–1985 and 1985–2003. There was a conflict between the new growth regime based on export and old institutional changes shaped for domestic consumption growth, and this deepened the lack of complementarity in the economy. In other words, although the country began pursuing export growth with open economic policies and relatively deregulatory policies, institutional factors remained largely unchanged. For instance, the trade unions maintained their position, privatizations were not implemented to any great extent and the government continued its populist income policies. Thus, in the period 2000–2001, the inconsistency between the new growth regime and old institutional factors caused the biggest economic crisis. This problem manifested itself because of uncompleted regulations that should have created institutional complementarity and eliminated path dependency remained from pre-1980s. Compared with previous periods, the mode of employment changed visibly. The role of manufacturing workers became dominant in the economy and exports of manufacturing products in total export passed those of agricultural products in the early 1980s. The relatively developed manufacturing industries in the closed and protected economic period began to compete in international trade. However, the position of the trade unions remained strong. Regarding wage-labor relations, the pressure of trade unions on government policies increased significantly. Spurred by expectations of institutional change and hardening economic conditions compared with the previous periods, the number of strikes peaked (see Table6). Strikes and lock-outs increased more rapidly than in the 1960s and 1970s. Workdays lost to these industrial actions by both trade unions and employer unions reached their highest level in the 1980s. In addition, wage rate growth increased significantly more than in previous periods. This economic unrest made Turkey follow a different path than the developed countries in the 1980s. Turkey attempted to implement a number of institutional changes to eliminate the increasing pressure from workers; changes that influenced the employment structure of the manufacturing sector. Neoliberal policies combined with high productivity growth of exports goods and that created new strategies of Turkey according to the new growth regime. Although export growth should have been associated with lower cost production in international trade to maximize gain, Turkey experienced the highest ULC growth of export goods in its history, which wrecked the fixed exchange rate system and led to devaluations. Thus, in this period, the exchange rate system was transformed from a fixed to a managed system. Nevertheless, that did not stop devaluations in the lira. Contrary to the previous period, the lira became an excessively inconsistent and overvalued currency. The change rate in the lira against the US dollar was −59.3%, whereas that of PPP was −64.3% in the period 1985–2003 (see Table2). The lack of institutional complementarity between macroeconomic and institutional factors for stable export Page 31 of 38 Ünal Economic Structures (2018) 7:3 Table 8 Periods of transformation in the Turkish economy Category 1950–1962 1962–1985 1985–2003 Post-2003 Macroeconomic factors Growth strategy Based on agricultural production Based on domestic consumption Transformation period and conflict between old institutional factors and new growth regime Based on export growth Wage growth Relatively low High High Relatively low Productivity Based on agriculture Based on non-tradable goods Based on export goods Based on export goods Inflation Low High High Relative low based on inflation targeting Trade balance Deficit Deficit Deficit Large deficit Institutional factors Employment Agricultural workers Agricultural and manufacturing workers Manufacturing workers Manufacturing workers Trade Union No Strong Strong Relatively strong Exchange rate Fixed Fixed and under-valued Managed and over-valued Floating and over-valued Production base Agricultural sector Protected domestic manufacturing Export goods sector Export goods sector Privatization No No Relatively low Large Mode of economy Open and liberal economy Protected economy Open economy Excessively open economy Government’s intervention Strong Strong Strong Relatively weak Role of Government Supports agricultural sector Supports domestic manufacturing Supports export goods sector Supports export goods sector (excessively private) Page 32 of 38 Ünal Economic Structures (2018) 7:3 Acknowledgements Thanks go out to Robert Charles Perry for his input and efforts on this paper. Competing interests The author declares that he has no competing interests. Availability of data and materials Various sources were used to collect data for the analyses. Each source of data and materials has been available and referred to throughout the paper wherever appropriate. Consent for publication Not applicable. Ethics approval and consent to participate Not applicable. Funding Not applicable. Appendix Calculating productivity growth of export goods and non-tradable goods In this equation, A is the technological coefficients’ matrix, y represents the vector of final demand and x is a vector that shows the level of output. The Leontief inverse matrix (I–A)−1 is used to calculate the labor required to directly and indirectly produce one unit of each commodity. To measure the productivity growth of non-tradable goods and export goods, the equation below is used: In Eq.(7), x is a column vector that shows the total amount of output for each commodity, A is the input coefficient matrix, which shows the amount of domestic commodities used by the industry to obtain one unit of output. Furthermore, φ is a row vector that shows the amount of labor that is directly used to produce one unit of output in each industry. Finally, L is a scalar that shows the total labor on the input–output table. where v is a row vector whose elements show the amount of labor that is directly and indirectly required to produce one physical unit of each commodity. The amount of total domestic final demand is indicated by N, and the amount of total exports is indicated by E. Furthermore, the shares of each commodity in this total are indicated as column vectors n and e, respectively. (5) Ax +y=x (6) y=x−Ax x =( I − A )−1 y (7) y=(I−A)x φx = L (8) φ(I − A)−1 = v (9) vy =v(N+E)=L (10) v n=  k=1 vknkand ve=  k=1 vkek Page 33 of 38 Ünal Economic Structures (2018) 7:3 In the equation, vn and ve are the vertically integrated labor input coefficients of nontradable goods and export goods, respectively. Labor productivity is calculated by means of vertically integrated input labor coefficients in each factor in both demand and exports.32 These coefficients are multiplied with price deflators.33 If the coefficients decrease, the productivity of non-tradable goods and export goods increases. Figures6, 7, 8 and Tables9, 10, 11, 12 mentioned through the paper are inserted as follow: 32 For additional information, see Ünal (2016a). 33 Deflators were derived and calculated from the UN database using “national accounts estimates of main aggregates” and “GDP by type of expenditure” categories. Deflators of Turkey were calculated from data derived from TurkStat (GDP in chain linked volume, 2009=100), national currency was converted into US dollars.ve deflator is calculated fromexportsat current prices (US dollars) divided by exportsat constant prices (US dollars).vn deflator is calculated from domestic demandat current prices (US dollars) divided by domestic demandat constant prices (US dollars). The following equation was used: Domestic demand=GDP−export+import. Page 34 of 38 Ünal Economic Structures (2018) 7:3 .0 .1 .2 .3 .4 .5 .6 .7 .8 5.6 5.7 5.8 5.9 6.0 6.1 6.2 .0 .1 .2 .3 .4 .5 .6 .7 .8 5.5 5.6 5.7 5.8 5.9 6.0 6.1 6.2 6.3 6.4 Canad aF rance .0 .1 .2 .3 .4 .5 .6 .7 5.5 5.6 5.7 5.8 5.9 6.0 6.1 6.2 6.3 6.4 6.5 6.6 6.7 .0 .1 .2 .3 .4 .5 .6 .7 .8 5.2 5.3 5.4 5.5 5.6 5.7 5.8 5.9 Japan Netherlands .0 .1 .2 .3 .4 .5 .6 .7 .8 5.5 5.6 5.7 5.8 5.9 6.0 6.1 6.2 6.3 .0 .1 .2 .3 .4 .5 .6 .7 .8 .9 6.3 6.4 6.5 6.6 6.7 6.8 6.9 7.0 7.1 7.2 United Kingdom United States Fig. 6 Sectoral shares of value-added in the three sectors, and GDP in developed countries. Note: Horizontal line is Log (real GDP). Source: Author’s calculations. Various sources were used to collect data. For Canada, UN data (national currency) was used between 1970 and 2014. For other countries, value-added at national prices (2005 = 100) were derived from Groningen Growth and Development Centre: For France the period 1950–2009; for Japan the period 1953–2011; for Netherlands the period 1960–2009; for the United Kingdom the period 1949–2009; for the United States 1950–2010. For additional information see Fig. 1. Page 35 of 38 Ünal Economic Structures (2018) 7:3 .0 .1 .2 .3 .4 .5 .6 .7 7.8 7.9 8.0 8.1 8.2 8.3 8.4 8.5 8.6 8.7 Log Agriculture Manufacturing Service (real GDP) Fig. 7 Sectoral shares of value-added in the three sectors, and GDP in Turkey (annually, 1970–2015). Source: Author’s calculation. Value-added (2005 = 100) was derived from UN data (US dollars). For additional information see Fig. 2 .0 .1 .2 .3 .4 .5 .6 .7 6.8 7.0 7.2 7.4 7.6 7.8 8.0 8.2 8.4 8.6 8.8 Log(real GDP) Agriculture Manufacturing Service Fig. 8 Sectoral shares of production in the three sectors, and GDP in Turkey (annually, 1923–2014). Source: Author’s calculation. Production is real value in each sector, derived from the Ministry of Development’s report (Economic and Social Indicators between 1950 and 2014). For additional information see Fig. 2 Page 36 of 38 Ünal Economic Structures (2018) 7:3 Table 9 Approximate sectoral wage growth in the countries (annual rate, unit: %). Source: Author’s calculations. For additional information see Table 1 Country Period Non-tradable goods Export goods Canada 1971–1981 9.9 11.2 1981–1990 5.2 5.3 France 1972–1980 14.6 13.4 1980–1990 7.9 8.3 Japan 1970–1980 11.7 11.8 1980–1990 7.5 7.0 Netherlands 1972–1981 9.9 10.4 1981–1986 6.5 6.2 Turkey 1973–1985 37.4 38.7 1985–2003 59.8 63.0 2003–2011 10.5 11.2 United Kingdom 1968–1979 10.8 11.5 1979–1990 8.7 9.0 United States 1972–1977 8.2 8.5 1977–1990 7.0 8.5 Table 10 Sectoral shares of production in developed countries (annual rate, unit: %). Source: Author’s calculations. Data were derived from OECD input–output tables. The data 1995 were derived from WIOD for Japan and Netherlands The periods when the structural transformation experienced were considered. The italic emphasis the period when the sectoral share of production in service sector surpassed that in manufacturing sector Country Period Agriculture Manufacturing Service Canada 1971 4.3 52.1 43.6 1981 5.0 52.9 42.1 1990 3.4 46.7 49.8 France 1972 7.7 57.5 34.8 1980 5.2 50.1 43.7 1990 4.0 43.0 53.0 Japan 1970 4.5 62.6 32.8 1980 3.0 57.7 39.3 1990 2.1 52.8 45.1 1995 1.7 45.6 52.6 Netherlands 1972 5.9 53.0 41.1 1981 4.8 52.4 42.7 1986 4.9 49.7 45.3 1995 2.2 38.5 59.3 United Kingdom 1968 3.4 59.5 37.1 1979 3.0 56.6 40.3 1990 2.0 44.7 53.3 United States 1972 4.2 47.9 47.8 1977 3.8 49.1 47.0 1990 2.6 40.4 57.0 Page 37 of 38 Ünal Economic Structures (2018) 7:3 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Received: 6 March 2017 Accepted: 27 October 2017 References Aboites J, Miotti L, Quenan C (2002) Regulationist approaches and accumulation in Latin America. In: Boyer R, Saillard Y (eds) Régulation theory: the state of the art (trans: Shread C). Routledge, London, pp 280–288 Balassa B (1964) The purchasing-power parity doctrine: a reappraisal. J Polit Econ 72(6):584–596 Boyer R (1990) The regulation school: a critical introduction (trans: Charney C). Columbia University Press, New York Boyer R, Hollingsworth JR (1997) The variety of institutional arrangements and their complementarity in modern economies. In: Boyer R, Hollingsworth JR (eds) Contemporary capitalism: the embeddedness of institutions. Cambridge University Press, Cambridge, pp 49–54 Boyer R, Saillard Y (2002) A summary of regulation theory. In: Boyer R, Saillard Y (eds) Regulation theory: the state of the art (trans: Shread C). Routledge, London, pp 36–44 Boyer R, Yamada T (2000) Introduction: a puzzle for economic theories. In: Boyer R, Yamada T (eds) Japanese capitalism in crisis: a regulationist interpretation. Routledge, London, pp 1–16 Briones R, Felipe J (2013) Agriculture and structural transformation in developing Asia: review and outlook, No. 363, ADB Economics Working Paper Series, Asian Development Bank Herrendorf B, Rogerson R, Valentinyi A (2013) Growth and structural transformation, NBER working paper 18996, National Bureau of Economic Research Jessop B, Sum NL (2006a) Introduction. In: Jessop B, Sum NL (eds) Beyond the regulation approach: putting capitalist economies in their place. Edward Elgar Publishing, Cheltenham, pp 1–10 Jessop B, Sum NL (2006b) Fordism and post-fordism. In: Jessop B, Sum NL (eds) Beyond the regulation approach: putting capitalist economies in their place. Edward Elgar Publishing, Cheltenham, pp 35–90 Kaldor N (1961) Capital accumulation and economic growth. Reprinted from the theory of capital, MacMillan & CO LTD, pp 177–222 King JE (2009) Nicholas Kaldor. Palgrave Macmillan, Hampsphere, p 77 Kuznets S (1957) Quantitative aspects of the economic growth of nations: II. Industrial distribution of national product and labor force. Econ Dev Cult Change 5(4):1–111 Kuznets S (1973) Modern economic growth: findings and reflections. Am Econ Rev 63(3):247–258 Michaels G, Rauch F, Redding SJ (2012) Urbanization and structural transformation. Q J Econ 127(2):535–586 Nakakubo H (2015) Industrial action and liability in Japan: a legal overview. Jpn Labor Rev 12(2):86–105 Samuelson P (1964) Theoretical notes on trade problems. Rev Econ Stat 46(2):145–154 Table 11 Wage growth in sectors and inflation in Turkey (annual rate, unit: %). Source: Author’s calculation. Wage growth was calculated by deriving data from TurkStat and WIOD input–output tables (compensation of employees) Period Agriculture Manufacturing Service Inflation 1973–1985 38.6 35.3 35.3 38.5 1985–2003 63.9 65.9 68.1 63.5 2003–2009 10.8 12.7 15.6 11.5 Table 12 Ratio of export to import by sector in Turkey. Source: Author’s calculation. Ratios were calculated by deriving data from TurkStat Period Agriculture Manufacturing Service 1973 17.20 0.24 0.13 1985 5.90 0.57 0.13 2003 0.87 0.69 0.10 2011 0.60 0.58 0.07 2016 0.82 0.73 0.08 Page 38 of 38 Ünal Economic Structures (2018) 7:3 Solow RM (1970) Growth theory: an exposition. Clarendon Press, Oxford, pp 1–16 Thelen K (2014) Varieties of liberalization and the new politics of social solidarity. Cambridge University Press, New York, pp 154–192 Ünal E (2016a) A comparative analysis of export growth in Turkey and China through macroeconomic and institutional factors. Evolut Inst Econ Rev 13(1):57–91. https://doi.org/10.1007/s40844-016-0036-3 Ünal E (2016b) How can turkey be part of the EMU? Turkey’s economic and monetary integration into the EMU and the analyses of macroeconomic and institutional factors by export-led growth. Kyoto Econ Rev 85(1–2):2–41 Ünal E (2017, forthcoming) Turkey’s Current account deficit problem and integration into the economic and monetary union of the European Union. Kyoto Econ Rev 86(1–2) Uni H (2007) Export-biased productivity increase and exchange rate regime in east Asia. Kyoto Econ Rev 76(1):117–138 Uni H (2012) Comparative analysis of conditions for monetary integration: Europe and Asia. In: Boyer R, Uemura H, Isogai A (eds) Diversity and transformations of Asian capitalism. Routledge, London, pp 287–305 Verdoorn PJ (2003) Factors that determine the growth of labor productivity. In: McCombie J, Pugno M, Soro B (eds) Productivity growth and economic performance: essays on Verdoorn’s Law (trans: Thirlwall AP). Palgrave Macmillan, New York, pp 28–36