The present study attempts to re-investigate the behaviour of disaggregated public expenditures data and national income for Malaysia. This study covers the sample period of annual data from 1960 to 2007. The Bartlett-corrected trace tests proposed by Johansen (2002) were used to ascertain the presence of long run equilibrium relationship between public expenditures and national income. The results show one cointegrating vector for each specification of public expenditures. The relatively new MWALD test indicates a strong unidirectional causal effect runs from national income to public expenditures in Malaysia. While, bilateral causality evident exists merely between public expenditure on health and national income. In Malaysia, it is not a wise strategy to solely depend on fiscal policy for long-term economic development.