Private Capital Flows And Macroeconomic Performance: Growth Implications Of Shocks For Sub-Saharan African Economies

ABSTRACT In theory, private capital flows (PCF) strengthen domestic investment for economic growth. In sub-Saharan African (SSA) economies, Foreign Direct Investment per Capita (FDIC), Portfolio Investment per Capita (PIC) and Bank Lending per Capita (BLC) components of PCF grew inversely to Gross Domestic Product per Capita (GDPC). While growth rates of FDIC, PIC and BLC respectively averaged 269.6%, 31.7% and 55.6% in 1981-1990; 42.9%, 36.6% and 28.6% in 1991-2000; 30.7%, - 174.7% and 24.2% in 2001-2010; GDPC growth rates were -1.3%, -0.4% and 2.2% over the periods. Previous studies have attributed this problem largely to recipient economies’ structural features, with little attention paid to PCF shocks (sharp fluctuations from the equilibrium path). This study, therefore, investigated the effects of PCF shocks on the macroeconomic performance of selected SSA countries. A stochastic model within a dynamic open-economy framework was developed to evaluate the relationship between shocks to gross inflows of PCF components (FDIC, PIC and BLC) and macroeconomic performance indicators (GDPC, Gross Fixed Capital Formation per Capita (GFCC), and Exchange Rate (ER)). Shocks were measured, using the Structural Vector Autoregressive (SVAR) model, as one-standard deviation of orthogonal structural errors. The Maximum Likelihood estimation technique employed yielded asymptotically efficient estimators which were invariant to the model’s re-parameterisation. The effects of the shocks on GDPC long-term growth were determined using the instrumental variables regression method. Annual data on fourteen SSA countries from 1990 to 2010 were employed, based on data availability. The data were collected from the International Monetary Fund’s International Financial Statistics Yearbook and the World Bank’s Global Development Finance databases. Reliability and robustness of estimators were ascertained using Johansen-Fisher co-integration and SVAR stability tests. Statistical significance was determined at 0.05 level.