The twin deficits hypothesis asserts that a reduction in the budget deficit causes a reduction in the current account deficit. The Keynesian hypothesis proposes that the causality runs from budget deficits to current account deficits. However; conflicting theories have been proposed, arguing that possibility exists of reverse causality from current account to budget deficit and indeed that there is no relationship between the two deficits. Proponents of the Ricardian equivalence hypothesis (REH) suggest the absence of any relationship between the current account deficit and the budget deficit. This study uses the Autoregressive Distributed Lag method (ARDL) of co integration to test the three proposed hypotheses using annual time series data of Malawi over the period 1970 to 2012. Three separate models were constructed to test the theories outlined above. Results from the analysis found a positive significant long run relationship between the budget deficit and the current account deficit. Implying that in the long run budget deficit does influence the current account deficit, asserting that the Keynesian proposition holds in Malawi. GDP and the real exchange rate were found to have a positive long run impact on the current account, with the current account responding much higher to the GDP than real exchange rate. No evidence was found in support of the reverse causality or the Ricardian equivalence hypothesis.