Export Performance and Economic Growth Causality: An Empirical Analysis

Hateml-J, Abdulnasser, Irandoust, Manuchehr, Atlantic Economic Journal

MANUCHEHR IRANDOUST [*]

The cointegration and causal relationship between export growth and economic growth is investigated for the Nordic economies. On the basis of Johansen's technique and the augmented Granger causality tests, the evidence shows that these macroeconomic aggregates are causally related in the long run for each economy. Granger causality is unidirectional, running from economic growth to export growth in Denmark, and bidirectional in Finland, Norway, and Sweden. The established bidirectional causality suggests that the expansion of exports is an integral part of the economic growth process. (JEL F41, F43, C32)

Introduction

International trade theory suggests that export growth contributes positively to economic growth (measured by output growth) through such means as:

1) facilitating the exploitation of economies of scale for small open economies [Helpman and Krugman, 1985];

2) relieving the binding foreign exchange constraint to allow increases in imports of capital goods and intermediate goods [McKinnon, 1964];

3) enhancing efficiency through increased competition [Balassa, 1978]; and

4) promoting the diffusion of technical knowledge, in the long run, through foreign buyers' suggestions and learning by doing [Grossman and Helpman, 1991].

Traditionally, it has been assumed that exports are exogenous to domestic output. This could be an inappropriate assumption because output can also affect exports. A justification of causality from output to exports may be found in Kaldor's [1967] contributions to the theory of growth. Kaldor shows that output growth has a positive impact on productivity growth, and improved productivity, or reduced unit costs, is expected to act as a stimulus to exports. From a policy-making point of view, it could be interesting to study the causal nexus of exports and output. Theoretically, the augmented production function has been used to show that export growth promotes economic growth [Krueger, 1978; Balassa, 1978; Greenaway and Sapsford, 1994]. In the augmented production function, real output depends on capital, labor, and other macroeconomic variables such as exports and industrial production. A positive correlation between export growth and real output growth is labeled in literature as the export-led growth hypot hesis, reflecting the view that export-oriented policies contribute to economic growth.

Does a long-run steady-state relationship exist between the key macroeconomic aggregates? Is there any comovement or causality between exports and output? These are the main questions addressed in this study concerning Denmark, Finland, Norway, and Sweden. Thus, the purpose of this paper is to investigate the causal nexus of export and output in the four Nordic economies.

Empirical research on the exports-output linkage has produced mixed results for the existence of any causal relationship between export growth and output growth. Among the studies that have supported the export-led growth hypothesis are Michaely [1977], Balassa [1978], Feder [1982], Chow [1987], Bahmani-Oskooee and Alse [1993], Thornton [1997], Doyle [1998], and Xu [1998]. Other studies, such as Granger [1969], Sims [1972], Jung and Marshall [1985], Hsiao [1987], and Afxentiou and Serletis [1991], have not supported the export-led growth hypothesis.

The problem with some of the previous studies (for example, Michaely [1977], Balassa [1978], and Feder [1982]) is that their conclusions are based on nonstationary data. In other words, very little consideration is given to evaluating the time series characteristics (stationarity and cointegration) of the data employed. The application of conventional econometric techniques to nonstationary (integrated) time series can give rise to misleading results and erroneous inference [Sims et al., 1990]. The assumption of stationarity and the imposition of a priori restrictions on the interrelationships among the variables create biased results because of the failure to investigate any mutual dependence among the variables. The question of whether or not to use level or difference data should be carefully examined subsequently in estimation and inference. Furthermore, with few exceptions, previous studies suffer from a major shortcoming, that is, the arbitrariness in the choice of the lags. The choice of the lags and the characteristics of the data are quite important and should be handled properly in causality tests. This is because causality tests are sensitive to the presence of unit roots and to model selection [Sims et al., 1990].

Additionally, a number of previous studies have focused mainly on estimating models using cross-sectional data based on the implicit assumption that countries have common characteristics. This may not be true …

Questia, a part of Gale, Cengage Learning. www.questia.com
Publication information: Article title: Export Performance and Economic Growth Causality: An Empirical Analysis. Contributors: Hateml-J, Abdulnasser - Author, Irandoust, Manuchehr - Author. Journal title: Atlantic Economic Journal. Volume: 28. Issue: 4 Publication date: December 2000. Page number: 412. © 1999 Atlantic Economic Society. COPYRIGHT 2000 Gale Group.

This material is protected by copyright and, with the exception of fair use, may not be further copied, distributed or transmitted in any form or by any means.