Some Observations on the Convergence Experience of Turkey

Abstract

We explore some aspects of the convergence process of Turkey and provide some international comparisons tracking the changes in per capita income figures. Turkey lost ground in the last few decades (1977–2001) of the twentieth century. After the lost decades, Turkey had high growth rates during 2002–2007. We argue that the rate of productivity growth will be an important factor for Turkey to converge to per capita income levels in developed countries. Our investigation with different countries and country groups provides motivation for rendering possible comparisons of the similarities between Turkey and Latin America.

Appendix A

Countries in Figure 1(a)

Albania, Algeria, Angola, Argentina, Australia, Austria, Bahrain, Bangladesh, Barbados, Belgium, Bolivia, Brazil, Bulgaria, Burkina Faso, Cambodia, Cameroon, Canada, Chile, China, Colombia, Costa Rica, Côte d’Ivoire, Cyprus, Denmark, Dominican Republic, DR Congo, Ecuador, Egypt, Ethiopia, Finland, France, Germany, Ghana, Greece, Guatemala, Hong Kong, Iceland, India, Indonesia, Iran, Iraq, Ireland, Israel, Italy, Jamaica, Japan, Jordan, Kenya, Luxembourg, Madagascar, Malawi, Malaysia, Mali, Malta, Mexico, Morocco, Mozambique, the Netherlands, New Zealand, Niger, Nigeria, Norway, Oman, Pakistan, Peru, Philippines, Poland, Portugal, Romania, Saudi Arabia, Senegal, Singapore, South Africa, South Korea, Spain, Sri Lanka, St. Lucia, Sudan, Sweden, Switzerland, Syria, Taiwan, Tanzania, Thailand, Trinidad and Tobago, Tunisia, Turkey, Uganda, the United Kingdom, the United States, Uruguay, Venezuela, Vietnam, Yemen, Zambia, and Zimbabwe. Countries in Figure 1(a) are selected such that there is no missing observation in the sample. Some countries such as former Soviet Union countries and some oil-rich Arab countries (Kuwait, Qatar, and the United Arab Emirates) are excluded.

Growth accounting: Some comparisons

Alternative decompositions

We present two alternative frameworks following Hsieh and Klenow (2010). Dividing equation 1 by working-age population (N) and rearranging yields the following expression:

An alternative accounting is based on the following rearrangement:

Hsieh and Klenow (2010) provide a discussion for both equations. We describe the results of the accounting exercises using equations A1 and A2. We use the same data as in the section ‘Sources of growth in the post-2001 period.’ In addition, data for working-age person (N), (population 15 years old and over) are from the Ministry of Development of Turkey, Economic and Social Indicators (1950–2010). Table A.1 presents the results for the 2002–2007 period. The larger exponents on TFP [1/(1−α) instead of 1] and on effective labor [1 rather than (1−α)] in equation A2 reflect the impact these two variables on output both directly and indirectly through capital per worker. In our case, α=0.5. Therefore, the last figure in Table A.1 is 5.18 (=2*2.59). The exercise based on Equation A2 implies that over the period 2002–2007, GDP per economically active person grew at 4.85% per year, which is completely accounted for by a 5.18% growth rate in TFP factor, A^1/(1−α).

Table A.1 Alternative decompositions (average annual changes, in percentage)

A Comparison of TFP growth rates

Figure A.1 compares the annual growth rates of our calculated TFP series with the TFP growth reported by the Total Economy Database (TED) during 1999–2010 for Turkey.⁹ We compute the annual growth rate in our calculated TFP series with the formula g _t =100*[(TFP _t /TFP_t−1)−1], where t denotes the year and g _t denotes the TFP growth rate for year t.

Figure A1

TFP growth in Turkey (in percentage), 1999–2010.

Source: Author’s calculations and the Conference Board Total Economy Database, January 2014

We observe that our calculations are comparable with the TED and the simple correlation between these two series is 0.99, indicating similarities. We calculate that TFP growth in 2010 is 3.48% and the corresponding figure reported by the TED is 3.36% for 2010. On the other hand, there are some differences between 2002 and 2007. For example, the average growth rate in TFP, during 2002–2007, reported in the TED is 1.43%, whereas our calculations suggest that the average TFP growth rate in the same period is 3.24%.

A comparison with China

What is the relative performance of Turkey, in the high-growth era of 2002–2007, compared to China, the leading emerging economy? We replicate the growth-accounting exercise for China. All data for China (except human capital) are from Kehoe and Ruhl (2010). Human capital is constructed following the procedure explained in the section ‘Sources of growth in Turkey’ and α=0.5. This value of capital income share is also consistent with China’s average capital income share as reported in China’s national accounts (see Zhu, 2012).

Table A.2 shows the sources of growth in output per worker for China during 2002–2007. The results presented in Tables 3 and A.2 are based on the national currency estimates. Therefore, the results are not directly comparable. However, the figures are useful to see the growth performance of Turkey and China, since they provide information on average annual growth rates. TFP growth is the dominant factor behind the Chinese growth. This is not surprising, since the recent literature on China comes to the conclusion that aggregate productivity growth has been the most important source of China’s growth since 1978 (see Zhu, 2012 and the references therein).

Table A.2 Sources of growth in China (average annual changes, in percentage)