Purpose – This paper investigates the twin deficit hypothesis for Ghana in view of the persistent co-movement between the budget deficit (BD) and current account deficit (CAD) over the past three decades. Design/methodology/approach – The paper uses annual data for the period 1980-2014 and ascertains whether there is long-run association between the budget and current account deficits using the Johansen Cointegration test. The Error Correction Model is estimated to check stability of the long-run association between the two deficits. A Granger causality test is performed to determine the direction of causality between the two variables. Findings – The results confirm the existence of long-run equilibrium relationship between the budget and current account balances. The error correction model finds an insignificant effect of the BD on the CAD both in the short and long runs. The ECM result was however significant for both the long run and short run regarding the effect of CAD on BD even as the adjustment parameter suggests that 33 percent of the disequilibrium in budget balance in the previous period is corrected in the current period. Granger causality results find support for the reverse causality argument, thereby rejecting the twin-deficits hypothesis for Ghana. Research limitations/implications – For lack of consistent data variables such as exchange rate and interest rate were not included in the study. Originality/value – The findings lend support to the reverse causality argument that says that causality runs from the current account deficit to the budget deficit.