Do US Federal Revenues and Expenditures Respond Asymmetrically to Budgetary Disequilibria?

The basic answer to the question posed by the title is: yes. We follow Ewing et al. (2006) and examine the US federal revenue-expenditure nexus in an error-correction model allowing for asymmetric adjustment. Symmetric adjustment is rejected by data from the 1959.3 to 2007.4 period. However, as do Ewing et al., I employ both threshold autoregression (TAR) and momentum threshold autoregression (M-TAR) models; depending on whether the TAR model or M-TAR model is employed, the estimated asymmetry is very different. The M-TAR model suggests that that revenues and expenditures respond to budgetary disequilibria only when the budget is worsening while the TAR model suggests responses only when the budget is improving. Ewing et al. take a strong stand in preferring the M-TAR-based results. I, however, argue that there is little justification for preferring one model over the other. More importantly, I also argue that the concepts of “worsening” or “improving” budgets are model-contingent in important way. Taking this into account, the results across models need not be contradictory.