Exploring the Determinants of Business Performance: A Regional Perspective Using Spatial Econometrics and AI-Powered Analytics

The aim of the present paper is to analyse the determinants of the performance (measured by total factor productivity TFP) of 309 Moroccan manufacturing firms based on the World Bank Enterprise Survey’s data (WBES) in 2023. The paper adopts a hybrid approach combining econometric methods, namely the multilevel model and the spatial autoregressive model (SAR), which helps analyse the joint impact of internal, regional and spatial factors, as well as artificial intelligence methods, the logarithmic K-means in particular, which enables the paper to deepen the analysis by clustering the firms sharing the same characteristics. The results reveal several findings. Regarding firm-specific characteristics, most of the variables have a positive and significant impact on firm performance, such as digitalisation, size, innovation and skilled workers; except for exports and age. Reasearch has shown that export has a non-significant impact while age has a non-linear effect. As for regional factors, the results of the multilevel model and the log K-means underline the importance of the regional framework on a firm’s productivity. Contrary to the expectations, the paper fails to confirm the impact of proximity on firm performance despite the inclusion of the SAR model. Future directions in this regard could base its findings on a more detailed regional analysis, which encompasses all Moroccan twelve regions along with the integration of data related to the distances between companies to estimate spatial effects more accurately.