Purpose - This paper introduces previously missing financial components(efficiency, activity and size) in the assessment of the finance-investment nexus. Design/methodology/approach - VAR models in the perspectives of VECM and short-run Granger causality are employed. Usage of optimally specified econometric methods in contradiction to purely discretionary model specifications in mainstream literature. Findings - Three main findings are established: (1) while finance led investment elasticities are positive, investment elasticities of finance are negative; (2)but for Guinea Bissau, Mozambique and Togo, finance does not seem to engender portfolio investment; (3)contrary to mainstream literature, financial efficiency appears to impact investment more than financial depth. Practical implications - Four policy implications result: (1)extreme caution is needed in the use of single equation analysis for economic forecasts; (2)financial development leads more to investment flows than the other way round; (3) financial allocation efficiency is more relevant as means to attracting investment flows than financial depth; (4) the somewhat heterogeneous character of the findings also point to shortcomings in blanket policies that are not contingent on country-specific trends in the finance-investment nexus. Originality/value - (1) Contrary to the mainstream approach we use four measures of financial intermediary development(depth, efficiency, activity and size) as well as four types of investment flows(domestic, foreign, portfolio and total). (2) The chosen investment and financial indicators are derived upon preliminary correlation analysis from the broadest macroeconomic dataset available on investment and financial intermediary flows.