2015˜The œJournal of developing areasRequires access

An empirical analysis of import demand function for Turkey: An ARDL bounds testing approach

Nazif Durmaz, Jaehyuk Lee

Open publisher page 23 citations

Abstract

Since the 1980’s, Turkish economy has revealed a dramatic growth. The purpose of the present study is to develop the import demand behavior of Turkey as an emerging economy and discover the long-run relation between the import demand and the components of GDP. One of the main reasons we adopt a disaggregated model to test aggregated import demand function estimation is that it deals with the potential aggregation bias when the different macro components have different import contents. The second reason is we believe that elimination of the biasedness will produce better forecasting results. This study examines the long-run and short-run elasticities of Turkey’s disaggregate import demand using an annual dataset that covers the period of 1980 to 2011. The error-correction modeling and cointegration analysis are the appropriate techniques since we try to discover the long-run relation between the import demand and components of GDP. First we determine the degree of integration of each variable in the model to apply the cointegration test that is known as an autoregressive distributed lag (ARDL). We collect the dataset from United Nations Statistical Division National Accounts Official Country Data and World Bank’s Development Indicators. This paper is one of the first studies to estimate import demand function of Turkey using the recent data which makes it different from earlier studies. Our results suggest that there is a long run relationship between the dependent variable and the control variables in the import demand function. All the explanatory variables are statistically significant in the long-run and the short-run as well. Our results also suggest that in the long run all of the included variables in the model are statistically significant. All the independent variables have inelastic effects on imports except total consumption. Relative price variable has a positive effect which may suggest in the short run importers cannot adjust price changes faster, consequently causing an increase in import expenditure bills. Thus, Turkish trade balance may get worsen if imports exceeds exports.

About this research paper

What this paper is about

Since the 1980’s, Turkish economy has revealed a dramatic growth. The purpose of the present study is to develop the import demand behavior of Turkey as an emerging economy and discover the long-run relation between the import demand and the components of GDP. One of the main reasons we adopt a disaggregated model to test aggregated import demand function estimation is that it deals with the potential aggregation bias when the different macro components have different import contents. The second reason is we believe that elimination of the biasedness will produce better forecasting results. This study examines the long-run and short-run elasticities of Turkey’s disaggregate import demand using an annual dataset that covers the period of 1980 to 2011. The error-correction modeling and cointegration analysis are the appropriate techniques since we try to discover the long-run relation between the import demand and components of GDP. First we determine the degree of integration of each variable in the model to apply the cointegration test that is known as an autoregressive distributed lag (ARDL). We collect the dataset from United Nations Statistical Division National Accounts Official Country Data and World Bank’s Development Indicators. This paper is one of the first studies to estimate import demand function of Turkey using the recent data which makes it different from earlier studies. Our results suggest that there is a long run relationship between the dependent variable and the control variables in the import demand function. All the explanatory variables are statistically significant in the long-run and the short-run as well. Our results also suggest that in the long run all of the included variables in the model are statistically significant. All the independent variables have inelastic effects on imports except total consumption. Relative price variable has a positive effect which may suggest in the short run importers cannot adjust price changes faster, consequently causing an increase in import expenditure bills. Thus, Turkish trade balance may get worsen if imports exceeds exports.

Why it matters

OpenAlex reports 23 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Since the 1980’s, Turkish economy has revealed a dramatic growth. The purpose of the present study is to develop the import demand behavior of Turkey as an emerging economy and discover the long-run relation between the import demand and the components of GDP. One of the main reasons we adopt a disaggregated model to test aggregated import demand function estimation is that it deals with the potential aggregation bias when the different macro components have different import contents. The second reason is we believe that elimination of the biasedness will produce better forecasting results. This study examines the long-run and short-run elasticities of Turkey’s disaggregate import demand using an annual dataset that covers the period of 1980 to 2011. The error-correction modeling and cointegration analysis are the appropriate techniques since we try to discover the long-run relation between the import demand and components of GDP. First we determine the degree of integration of each variable in the model to apply the cointegration test that is known as an autoregressive distributed lag (ARDL). We collect the dataset from United Nations Statistical Division National Accounts Official Country Data and World Bank’s Development Indicators. This paper is one of the first studies to estimate import demand function of Turkey using the recent data which makes it different from earlier studies. Our results suggest that there is a long run relationship between the dependent variable and the control variables in the import demand function. All the explanatory variables are statistically significant in the long-run and the short-run as well. Our results also suggest that in the long run all of the included variables in the model are statistically significant. All the independent variables have inelastic effects on imports except total consumption. Relative price variable has a positive effect which may suggest in the short run importers cannot adjust price changes faster, consequently causing an increase in import expenditure bills. Thus, Turkish trade balance may get worsen if imports exceeds exports.

Key concepts: Cointegration, Distributed lag, Economics, Econometrics, Demand curve, Error correction model, Variable (mathematics), Short run

Related papers

Back to paper searchBrowse research topicsOriginal source
An empirical analysis of import demand function for Turkey: An ARDL bounds testing approach — Research Paper | ScholarLens