Economics Working Paper 15117
This paper examines the use of hybrid economic models in the design and evaluation of multiyear fiscal consolidation strategies. In the United States, the Congressional Budget Office (CBO) has the responsibility for estimating the impact of such strategies. The CBO uses one type of model to estimate the short-run impact of the policy and another type of model to estimate the long-run impact. The two impacts are then spliced together in an ad hoc way. In principle it would be better to use one complete model—a hybrid model—in which the short-run and longrun decisions of people interact according basic dynamic economic theory. The key question is how, and how well, such an approach can be used in practice. To address this question, two periods of U.S. history are examined. In each period hybrid models had a role in the design or evaluation of such strategies, and their predictions were consistent with the outcomes. The results show that large multi-year credible deficit reduction plans can have positive effects in the short run and the long-run, a result that differs from the splicing approach in which the short run effects are always negative even for gradual phased-in credible plans.