Received: 27 October 2022 | Revised: 8 March 2023 | Accepted: 13 March 2023 | Published online: 15 March 2023 RESEARCH ARTICLE Resource Curse in WAIFEM Member Countries: An Application of Seemingly Unrelated Regression Nasiru Inuwa* , Maryam Bello and Mohammed Bello Sani Gombe State University, Nigeria Abstract: Even though empirical evidence has shown that naturally endowed countries growth slower than their less naturally endowed counterparts, the scenario tagged as “resource curse hypothesis,” but it seems there are exceptions. Therefore, this study examines the confirmation or disputation of resource curse hypothesis among the West African Institutes for Financial and Economic Management member countries during the period 1986–2016. The study applied seemingly unrelated regression and unraveled the strengthened effect of natural resources on output growth in Gambia, Ghana, and Sierra Leone. But, the study revealed a negative and statistically significant effect on economic growth in Liberia and Nigeria, thereby upholding the presence of resource curse hypothesis only in Liberia and Nigeria. Therefore, an overall “umbrella” policy recommendation inappropriate, but individually designed strategy that would help in managing the resource rents effectively in order to boost economic growth particularly in Liberia and Nigeria where their resource endowment serves as a curse rather than a blessing is recommended. Keywords: resource curse, economic growth, natural resources, SUR, WAIFEM 1. Introduction The role of natural resources in the growth and development of many nations has been at central axis of development theory and practice. Coincidently, a substantial number of developing countries have been endowed with different types of natural resources such as oil, natural gas, and other mineral resources. While judicious use of such resources might offer an opportunity to resource-abundant countries to transform their economies, only a handful of them were able to convert such endowments into sustained and meaningful growth (Thorborg & Blomqvist, 2015). Africa is generously endowed with not only renewable but also nonrenewable resources, and conventional economic theory posits that natural resources strengthened economic growth. Interestingly, some empirical studies justified that particularly Brunnschweiler (2008), Fan et al. (2012), and Wu et al. (2018). However, their findings to a larger extent have been challenged following the earlier studies of Sachs & Warner (1995), Papyrakis & Gerlagh (2007), Boschini et al. (2007), Daniele (2011), Behbudi et al. (2010), Eregha & Mesagan (2016), and Amini (2018) where detrimental effects of natural resources on economic growth were evidenced. This finding is tagged as “resource curse” in the natural resource economics literature. But a critical review of previous studies revealed many shortcomings. For instance, most of these studies (Atkinson & Hamilton, 2003; Gylfason 2001, 2002; Sachs & Warner, 1995, 1997) were based on a cross-sectional modeling using ordinary least square (OLS) methods, and their conclusions were highly unreliable and have been criticized based on different econometric problems. This is not farfetched as convincingly argued by Wijesekere (2009) that findings of studies modeled from the lens of cross-sectional framework are sensitive to number of samples chosen and wrongly treat different economic entities as homogeneous. Similarly, applying instrumental variables does not seem to have solved the problems especially when the long periods of data have been averaged. As for time series modeling, the solutions to the problems are yet to be obtained because larger observations are a prerequisite for robust econometric analysis, thereby restricting their analysis to a fewer number of countries where the data availability is another source of concern. Attempt to solve the problems identified with both cross- sectional and time series analysis led to introduction of panel data which combine the two modeling frameworks and offer a variety of estimation techniques such as static and dynamic panel techniques as well as panel cointegration tests. However, conventional fixed-effect and random-effect models produce biased and inconsistent estimates in the presence of endogeneity problem. Equally identified as a shortcoming is not only the inability to capture the dynamic nature of most growth models but also the homogeneity imposed across countries even when they differed at their developmental stages (Samargandi et al., 2014). Another point worth noting is the requirement of large numbers of cross-sections over time periods for the application of both difference and system generalized method of moments (GMM). Thus, failure to satisfy this condition might likely produce spurious results emanating from the number of instruments that will become larger and consequently affects the validity of Sargan *Corresponding author: Nasiru Inuwa, Gombe State University, Nigeria. Email: ninuwagsu@gmail.com Green and Low-Carbon Economy 2023, Vol. 00(00) 1–7 DOI: 10.47852/bonviewGLCE3202485 © The Author(s) 2023. Published by BON VIEW PUBLISHING PTE. LTD. This is an open access article under the CC BY License (https://creativecommons.org/ licenses/by/4.0/). 01