In the literature neither taxes, government spending nor deficits are robustly correlated with economic growth when evaluated individually. The lack of correlation may arise from the inability of any single budgetary component to fully capture the stance of fiscal policy. We use pair-wise combinations of fiscal indicators to assess the relationship between fiscal policy and U.S. growth. ; We develop a VAR methodology for evaluating simultaneous shocks to more than one variable and use it to examine the impulse responses for simultaneous, unexpected and equivalent structural shocks to pair-wise combinations of fiscal indicators. We also exploit the identity relationship between taxes, spending and deficits and follow Sims and Zha (1998) to evaluate an unexpected structural shock to one included fiscal indicator, holding constant the other included indicator. We find that an increase in the size of federal government leads to slower economic growth, that the deficit is an unreliable indicator of the stance of fiscal policy, and that tax revenues are the most consistent indicator of fiscal policy.