An extended Taylor rule for a small open African economy

Abstract

This paper estimated a monetary policy rule for Nigeria by augmenting the traditional Taylor rule model, using data spanning 2010Q1 to 2023Q2 in a Dynamic Ordinary Least Square regression. The results from competing models – the Taylor rule and extended Taylor rule (exchange rate augmented) – were subjected to forecast evaluation to ascertain their relative predictive power. Same models were estimated for South Africa for the purposes of comparison, and the results were consistent with Nigeria's. The findings indicate the predictors of the policy rate – monetary policy inertia, inflation, output gap and exchange rate – to be significant and with the expected signs. Forecast evaluation of the estimated extended Taylor rule model produced lower Root Mean Square Error (RMSE) and Theil statistics, compared with the traditional Taylor rule model, reflecting the higher predictive power of the former. The baseline and unequally-weighted two-quarter-ahead forecast of the extended model predicts a maximum inflation rate of 20.77 % and 21.04 % in 2023Q3 and 2023Q4, respectively, suggesting lower inflationary episode for Nigeria. The implication of the finding is that the extant monetary policy rule based on the traditional Taylor rule is inadequate in setting the nominal interest rate by the Central Bank of Nigeria, and recommends the adoption of the extended Taylor rule to enhance the effectiveness of monetary policy in Nigeria.