Non-linear Fiscal Effects of Uranium Price Shocks in Namibia
Description
The analysis focuses on three fiscal outcome indicators, each representing a distinct transmission channel through which critical mineral price shocks influence public finances (Table 1). The mining revenue share (LMRSH), measured as the natural logarithm of mining revenue relative to total government revenue, and the overall fiscal balance (FBAL), expressed as a percentage of GDP, are flow-based indicators that contemporaneously respond to commodity price movements. Therefore, these variables capture fiscal stability channels, reflecting short-run budgetary pressure and revenue volatility. By contrast, the public debt-to-GDP ratio (LDEBT), expressed in natural logarithms, is a stock variable that evolves cumulatively and reflects the government’s capacity to meet its debt obligations without an unsustainable fiscal adjustment. Thus, we interpret LDEBT as a fiscal sustainability indicator that captures the long-term consequences of asymmetric commodity price shocks. To ensure consistency across the model specifications, a common set of explanatory variables was included in all the estimations (Table 1). These include real GDP growth (GDPG), consumer price inflation (INF), the natural logarithm of mining value-added as a share of GDP (LMVA), the natural logarithm of the real effective exchange rate (LREER), and the natural logarithm of the real uranium price (LUP), which is deflated using the US CPI. In addition, a policy dummy variable (POLICY) is introduced, taking a value of one from 2021 onward to capture the implementation period of Namibia’s Minerals Beneficiation Strategy and zero otherwise.
Files
Steps to reproduce
Given the strong theoretical and empirical evidence of asymmetric fiscal responses to commodity price cycles, a linear specification is insufficient. The nonlinear autoregressive distributed lag (NARDL) framework is employed to allow positive and negative uranium price changes to exert distinct short- and long-run effects on fiscal stability and fiscal sustainability indicators. Bound testing is used to establish the existence of long-run relationships, while asymmetry tests evaluate whether these relationships are characterized by unequal effects of price increases and decreases. Granger causality tests are subsequently used to assess directional predictability, consistent with the proposed fiscal transmission mechanisms. The empirical analysis adopted a single-equation NARDL model estimated within an error-correction framework (ECM) separately for each dependent variable (LMRSH, FBAL, and LDEBT).
Institutions
- Namibia University of Science and TechnologyKhomas Region, Windhoek