Does Threshold Effect Exist in the Relationship Between Digital Intensity Index and Bank Efficiency in Nigeria?
Abstract
This study investigates whether threshold effects exist in the relationship between Digital Intensity Index (DII) and bank efficiency in Nigeria. While the growing literature on digital transformation suggests that digital technologies improve banking performance, empirical evidence remains inconclusive regarding whether such effects are linear or nonlinear. Using panel data from twelve Nigerian Deposit Money Banks (DMBs) covering the period 2000–2024, the study employs threshold regression techniques to examine the existence of critical digitalization levels beyond which the effect of digital intensity on efficiency changes significantly. Efficiency is measured using Data Envelopment Analysis (DEA) under both Window Technology and Constant Technology assumptions, as well as Stochastic Frontier Analysis (SFA). The findings reveal strong evidence of threshold effects. The DEA Window model identifies a threshold value of approximately 24.53, suggesting that banks begin to experience competitive efficiency gains at relatively low levels of digitalization. However, the DEA Constant Technology and SFA models identify substantially higher thresholds ranging from 66.76 to 72.64, indicating that structural efficiency gains emerge only after digital technologies become deeply embedded within banking operations. The results further reveal that digital intensity exhibits diminishing marginal effects on revenue efficiency and delayed positive effects on cost efficiency. These findings support the Productivity Paradox Theory, Dynamic Capability Theory, and Resource-Based View by demonstrating that the efficiency consequences of digital transformation are nonlinear, stage-dependent, and contingent upon the level of digital maturity attained by banks. The study contributes to the digital banking literature by proposing a dual-threshold framework comprising a competitive threshold and a structural threshold. The findings provide important policy implications for bank managers and regulators seeking to maximize the efficiency benefits of digital transformation within the Nigerian banking sector.
Keyword : Bank Efficiency, DEA, Digital Intensity Index, Digital Transformation, Nigeria Stochastic Frontier Analysis, Threshold Regression

This work is licensed under a Creative Commons Attribution 4.0 International License.