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

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 (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, 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.

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 development in advanced capitalist countries. Some of these facts are about the distinction features of labor productivity, capital and output in economy. He considered export-led 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.

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 all of these countries before the 1980s, and international trade was a secondary consideration. In other words, protected and regulated economies were dominant. When the sectoral shares of workers in the manufacturing sector started to decline toward or in the 1970s, domestic consumption growth was clearly visible (see Table 1). The same situation emerged in the sectoral share of value-added in the three sectors of the countries (see Fig. 6). Hence, it can be said that the development path of the developed countries changed in the 1970s and that drove workers into the service sector. In the 1980s, the sectoral share of workers in the manufacturing sector continued to decrease steadily, whereas in the service sector it increased. The gap between the sectoral shares of workers in the manufacturing and service sectors widened. The sectoral share of value-added in the two sectors showed the same pattern. In that period, the countries’ economies transformed to favor export growth. Thus, wage growth lost its importance as a stimulant for domestic consumption, and this helped inflation to decrease.

In Canada, between 1971 and 1976, the productivity growth of non-tradable goods was 1.2% higher than that of export goods. Wage rate growth was 11.0%, and inflation was 7.4%. High wage rate growth continued until 1980. However, between 1981 and 1986, the Canadian economy showed features of the Balassa–Samuelson model: the productivity growth of export goods increased significantly, to 5.4%, whereas the productivity growth of non-tradable goods was 0.9%. Also, inflation fell compared with previous periods, and the rate of change rate in its PPP and its currency decreased (see Table 2).

In France, between 1972 and 1977, the productivity growth of non-tradable goods, which was 6.0%, surpassed the productivity growth of export goods. In that period, wage rate growth was 13.5%—excessively high—and inflation was 9.5%. However, in the period 1985–1990, wage rate growth decreased significantly. Although it is not a perfect example, France had a relative tendency toward the low-cost production model because of its decreasing wage rate growth relative to the productivity growth of non-tradable goods, and its currency moved in parallel with its PPP (see Table 2). This shows that, in the 1980s, France abandoned domestic consumption growth and decreased wage growth to better equip for export growth.

Although Japan’s productivity growth of export goods was higher than the productivity growth of non-tradable goods, its wage rate growth was excessively high: approximately 13.3% between 1970 and 1980. Thus, inflation was high; around 8.9%. However, in following years, wage rate growth decreased significantly and the low-cost production model became dominant between 1980 and 1990; its currency remained over-valued, but it moved in parallel with its PPP (see Table 2).

The Netherlands followed domestic consumption growth until 1981, but between 1981 and 1986, wage rate growth decreased significantly to 2.3%, and, most importantly, the productivity growth of export goods, which was 7.0%, was higher than that of non-tradable goods. Hence, the Netherlands’ economy functioned under the low-cost production model.

Between 1968 and 1979, the United Kingdom experienced domestic consumption growth. In the period 1979–1984, this country had high wage rate growth of 9.8%, surpassing the productivity growth of non-tradable goods and export goods, which were 1.0 and 4.8%, respectively. Also, inflation was 10.2%. Between 1984 and 1990, the United Kingdom was able to decrease its wage rate growth slightly to 7.2%, and the inflation rate fell to 4.9%.

In the period of 1972–1977, the United States had low productivity growth and high wage rate growth, which boosted its inflation. Between 1977 and 1982, wage rate growth increased slightly to 8.1% and inflation climbed to 9.2%; unwanted conditions for a US economy experiencing low productivity growth of non-tradable goods and export goods. After 1977, the United States became an export growth country, where the productivity growth of export goods became higher than that of non-tradable goods. Between 1982 and 1985, the United States decreased its wage rate growth to reflect the productivity growth of export goods, as discussed in the Balassa–Samuelson model. During the period 1985–1990, the United States transformed to the low-cost production model by fixing its wage rate growth to the productivity growth of non-tradable goods.

In the 1970s and 1980s, the institutional forms of the developed countries discussed above also showed a parallel movement with each other. In the 1970s, these countries’ manufacturing sectors maintained a large number of workers, and trade unions were stronger and more effective compared to the 1980s. As a result, industrial actions were excessive. The 1970s witnessed an important institutional change in the fixed exchange rate system, when the Bretton Woods System collapsed. In that period, in the developed countries, the production base was manufacturing and the mode of economy was more domestic consumption-oriented, except in Japan, which prioritized export policies (see Table 1).¹⁴ In the 1970s, government intervention into economies was strong. In particular, government policy was to create stability in price levels, although this was not done successfully. The role of government was to support domestic manufacturing, as agricultural production had already lost its dominant economic role in most of the developed countries. In the 1970s, the developed countries were ready to start deregulation policies mandated by neoliberalism, which in turn stimulated economic growth in the 1980s. Trade unions lost much of their power and neoliberalism became the main means of creating flexible labor and open and excessively liberal economies in most countries.

In the 1980s, a period of deregulation and globalization emerged that systematically transformed economies to embrace the free market. The institutional forms of the developed countries changed in general with deregulation policies that were necessary to curb inflation. The growth strategies of the countries came to be based on export growth. Wage growth decreased as the trade unions became weaker. Moreover, as the sectoral share of workers declined in the manufacturing sector, the trade unions lost power because of decreasing union density. Anti-inflationary policies grew in prominence. Hence, decreasing labor cost growth played a significant role in lowering inflation. The exchange rate system became more flexible.¹⁵ The countries’ production was based on the export goods sector. Privatization became the policy of governments.¹⁶ Privatization transactions increased, and this also played a role in decreasing the power of trade unions.¹⁷ The mode of economy in the developed countries became excessively open and oriented to international competition. Government intervention into the economy became weaker and the role of government was more oriented to support the export goods sector than it had been in the previous period. In conclusion, the developed countries basically experienced a technological transformation in the 1970s as workers started moving into the service sector, and the countries discussed in this section completed their transformation in the 1980s to export growth policy.

Figure 2 shows the relationship between the distribution of workers among the various sectors and their movements with GDP in Turkey. As seen, there is a parallel movement of workers in the manufacturing sector and service sectors over these years, which shows a transformation away from the agricultural sector. With this transformation, GDP and GDP per capita increased. For instance, GDP was 62.2 billion US dollars in 1975 and 800 billion US dollars in 2014. Although GDP per capita remained below that of higher income countries, it was 1555 US dollars in 1975 and 10,404 US dollars in 2014.¹⁸ The figure does not show a downward inclination in the relationship between the share of workers in the manufacturing sector and GDP, and the level of GDP per capita, which remains lower than in developed countries, implying that Turkey is still a developing country and keeps its dynamism in structural transformation via the manufacturing sector. Thus, the manufacturing sector’s share of workers shows a steadily increasing trend. In 1923, the agricultural sector’s share of workers was 89.9%, and in 1950, it was 84.3%, whereas that of the manufacturing sector was 4.3% in 1923 and 8.6% in 1950. In 2014, the distribution of workers showed a different configuration: 21, 27.9 and 51% in the agricultural, manufacturing and service sectors, respectively.

Table 3 shows the correlation between the parameters discussed in Fig. 2. As seen, the proportion of workers in the agricultural sector was negatively correlated with real GDP and the distribution of workers between the manufacturing and service sectors. The important point is to note that two other parameters, the manufacturing and service sectors, are positively correlated with real GDP and with each other. This shows that worker distribution has been increasing both in the manufacturing and service sectors. Moreover, sectoral shares of value-added and production in these two sectors, and GDP showed a parallel movement as discussed in Fig. 2 (see Figs. 7, 8). For instance, between 1970 and 2015, the value-added share of the agriculture sector decreased on average by 2.3%, and those of the manufacturing and service sectors increased on average by 0.8 and 0.3%, respectively. In addition, between 1923 and 2014, the production share of the agricultural sector decreased on average by 1.5%, whereas those of the manufacturing and agricultural sectors increased on average by 1.2 and 0.7%, respectively. Furthermore, the manufacturing sector’s share in export goods compared with that of the agricultural sector rose in parallel with GDP (see Fig. 5). The important point is to see here that the Turkish economy is still on its development path as the manufacturing sector still does not have a clear downward inclination.

The structural transformation over time created a different configuration in forms of wage growth, and the ratio of export to import in these sectors (see Tables 11, 12). In the period of 1973–1985, wage growth in the agricultural sector was closer to inflation and to those of other sectors, but over the years, wage growth in the manufacturing and service sectors increased more than in the agricultural sector. In the period 2003–2009, inflation was 11.5%, and wage growth was 10.8% in the agricultural sector but 12.7% in manufacturing and 15.6% in the service sector. In addition, in 1973, the ratio of export to import was 17.2 in the agricultural sector and 0.24 in the manufacturing sector. As industrialization became dominant and agriculture lost its importance, the ratio decreased to 0.82 in the agricultural sector, but it increased to 0.73 in the manufacturing sector in 2016.

In this paper, stylized facts of structural transformation, macroeconomic and institutional factors in Turkey have been discussed by incorporating growth models via input–output analysis. Turkey experienced two significant technological changes. First, the country transformed from one of agricultural growth to one of manufacturing growth. Second, the country left domestic consumption growth and became an export growth country. To make the process clear, the periods of transformation regarding institutional factors were analyzed between 1950 and post-2003. In the 1950s, Turkey followed agricultural growth strategies, promoted by government intervention based on liberal economic policies. This economic transformation brought mechanization, which was stimulated by the import of goods. Hence, Turkey created significant economic growth in the agricultural industry. The agricultural growth strategies developed manufacturing, and this process was supported by domestic consumption growth strategies within a protected and closed economy, created following state planned organization which prepared 5-year development plans in the 1960s and 1970s.

Between 1962 and 1985, the Turkish economy was based on domestic consumption growth that naturally brought about legislation for trade unions’ rights of demonstration, strikes, and for employer unions, the right of lock-outs. The trade unions’ industrial actions combined with radical left wing movements against the government that made decisions in the early 1980s to follow open economic policies and abandon protectionism, but these decisions failed to bring rapid success. However, the country continued implementing institutional changes according to export growth. In the period 1985–2003, the transformation of the economy was slowed down because of political and economic problems. Hence, Turkey became a country under the over-valued currency model due to its repressed exchange rate and high-cost production, which is defined in this work as a long transformation period until the end of the 2000–2001 economic crisis.

In the 2000–2001 economic crisis, the government began its strict transition program via privatizations and important institutional changes aimed at decreasing inflation. Thus, the country was able to create relatively consistent export growth compared with that of previous periods. Nevertheless, compared with the macroeconomic factors of the developed countries, which could implement successive deregulation policies and minimized ULC growth, Turkey still could not create an institutional complementarity between its macroeconomic and institutional factors, and it remains under the over-valued currency model, which increases its economic vulnerability in terms of inflation and the exchange rate, and still affects its economic performance today (see Table 8).