John R. Graham (Duke, Fuqua School of Business) & Alan L. Tucker (Pace, Lubin School of Business) have posted Tax Shelters and Corporate Debt Policy. Here is the abstract:

Using a novel sample of 44 tax shelter cases involving public corporations, we investigate which types of firms shelter, the magnitude of the tax shelters they use, and whether participating in a shelter affects corporate debt policy. The propensity to shelter increases when a firm is large, profitable, has foreign operations, or large R&D expenditures, and decreases if the firm is audited by a Big 8 firm. The average deduction produced by the shelters in our sample is very large, equaling approximately nine percent of asset value. This is about three times as large as interest deductions for similar-sized, same industry firms. Seven years before they engage in sheltering activity, shelter firms have mean debt ratios of about 25 percent, roughly equivalent to similar-sized same-industry firms. By the year of the sheltering activity, shelter firm debt ratios have fallen to approximately 18 percent while matched firm debt ratios have not fallen. Our results suggest that corporations substitute away from debt when using tax shelters. These results help explain why some firms appear to be under-levered when tax-sheltering activity is ignored, and why corporate tax payments have fallen so precipitously in recent years.