The unintegrated economics of corporate tax integration

NOTE: This paper is an unfinished work of William D. Andrews, who was the Eli Goldston Professor of Law at Harvard Law School and a premier tax scholar of his generation. This substantial but not fully completed manuscript was found among his papers after his death on May 20, 2017. It examines significant details of the economic analysis contained in the 1992 Treasury Corporate Integration Report. Although that report is now more than twenty-five years old and this draft was written nearly two decades ago, the problems Andrews identifies in the report remain relevant today. We have chosen to publish the work unedited except for minor formatting changes (due in part to the conversion of a photocopy to Word) and a handful of minor corrections. Bill did not write an abstract, so the one appearing below is our description of his paper. This paper examines the economic analysis of the effects of corporate integration contained in the 1992 Treasury report on corporate integration. The core argument is that the central computable general equilibrium (CGE) models used to estimate the effects of the corporate tax (and the benefits of integration) assume free flows of funds between the corporate and non-corporate sectors whereas the Traditional View arguments for integration that are relied on by the Treasury assume the opposite, that the dividend tax imposes a charge on shifting funds out of corporate solution. After examining the structure of the CGE models in detail (using the Harberger two-sector model as an example), the paper shows how the assumptions made in these models are inconsistent with the assumptions made in the models of the tax effects of the financing decisions of corporations. The paper then considers some ways of potentially reconciling the two approaches