Non-linearity and the threshold effect of transparency on economic growth: evidence from developing countries

ABSTRACT

This study examines the linear and non-linear threshold impacts of transparency on economic growth in 70 developing countries from 1984 through 2018, utilizing panel generalized method of moments (GMM) and dynamic-panel threshold estimations. We find that the impact of transparency on growth is stronger when we use non-linear GMM, and dynamic-panel threshold estimations. There is a U-shaped relationship between transparency and economic growth, while transparency is found to stimulate economic growth, especially in a high-transparency threshold regime. The result reveals that financial development plays an important role in moderating the transparency–economic growth nexus in developing countries.

KEYWORDS:

• Transparency
• economic growth
• GMM estimation
• dynamic-panel threshold regression

JEL CLASSIFICATION:

• O11
• O43
• O47

Acknowledgements

We thank the anonymous reviewers and editor for useful comments and suggestions.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/17487870.2022.2136175

Notes

1. The original sample contained 82 developing countries. However, we eliminated 12 countries because they were considered as outliers using the Cook’s distance outlier test. Table A in the appendix lists the countries used in this study.

2. Numerous studies have used the same approach to construct the index based on ICRG datasets, including Law and Singh (2014), Law, Tan, and Azman-Saini (2015), Slesman, Baharumshah, and Ra’ees (2015), Ahmad and Hall (2017), and Vianna and Mollick (2018).

3. Overall, the indicators use 29 variables collected from 21 sources. The index is constructed based on the World Bank’s World Governance Index (WGI); a version of the unobserved-component model is used in which each source is classified as a “noisy” signal in the overall governance sub-category.

4. We would like to thank Stephanie Kremer for sending her Matlab code for the panel dynamic threshold estimations.

Additional information

Funding

This research was funded by the School of Business and Management, Institut Teknologi Bandung 139/I1.C12/SK/KP/2020 under the International Research Collaboration 2020.