Impact of Structural Breaks on the Fiscal Indicators–economic Growth Relationship in Nigeria: An ARDL and Bai-perron Approach

This study examines whether structural breaks alter the estimated relationship between fiscal indicators and economic growth in Nigeria. Annual data for 1981–2023 are analysed using the autoregressive distributed lag (ARDL) bounds-testing framework, while the Bai–Perron procedure identifies multiple structural breaks. Real gross domestic product represents economic growth, and the explanatory variables comprise total government expenditure, oil revenue, non-oil revenue, total federally collected revenue, Federation Account statutory allocation, Federal Government retained revenue, and total exports. The Bai–Perron test identifies break dates in 1998 and 2010. The baseline recursive CUSUM test indicates parameter instability before break dummies are introduced, whereas the model becomes stable after their inclusion. The bounds test supports a long-run relationship among the variables. Long-run estimates indicate positive and statistically significant effects of total government expenditure and Federation Account statutory allocation, while total exports have a negative and statistically significant effect. The 2010 break dummy is also positive and significant. The error-correction term is negative and significant, showing that approximately 17.16% of short-run disequilibrium is corrected in each period. Diagnostic testing detects serial correlation but not heteroskedasticity or non-normality; Newey–West heteroskedasticity- and autocorrelation-consistent standard errors are therefore used for robust inference. The findings indicate that recognising structural change improves parameter stability and the reliability of the estimated fiscal indicator–growth relationship in Nigeria.