Critical Finance Review > Vol 9 > Issue 1-2

Corporate Taxes and Capital Structure: A Long-Term Historical Perspective

Matthias Fleckenstein, University of Delaware, USA, mflecken@udel.edu , Francis A. Longstaff, UCLA Anderson School of Management and the NBER, USA, francis.longstaff@anderson.ucla.edu , Ilya A. Strebulaev, Graduate School of Business at Stanford University and the NBER, USA, istrebulaev@stanford.edu
 
Suggested Citation
Matthias Fleckenstein, Francis A. Longstaff and Ilya A. Strebulaev (2020), "Corporate Taxes and Capital Structure: A Long-Term Historical Perspective", Critical Finance Review: Vol. 9: No. 1-2, pp 1-28. http://dx.doi.org/10.1561/104.00000069

Publication Date: 11 Jun 2020
© 2020 Matthias Fleckenstein, Francis A. Longstaff and Ilya A. Strebulaev
 
Subjects
 
Keywords
G01G21G23H25
Corporate taxesLeverageCapital structure
 

Share

Download article
In this article:
1. Introduction 
2. The US Corporate Income Tax 
3. The Data 
4. How Do Corporate Tax Rates Affect Leverage? 
5. Robustness Results 
6. Conclusion 
7. Appendix 
References 

Abstract

We study the time series relation between leverage and corporate tax rates using an extensive data set constructed from all corporate income tax returns filed with the Internal Revenue Service from 1926 to 2013. This data set includes financial statement data from millions of private and public corporations. We find strong evidence that changes in corporate leverage are directly related to changes in corporate tax rates for all but the smallest firms. These results are robust to the inclusion of control variables for the costs of financial distress, corporate liquidity, and capital market and macroeconomic conditions. Furthermore, the results hold for both financial and nonfinancial firms as well as for different measures of firm leverage and the marginal corporate income tax rate. An increase in the marginal corporate tax rate of 1% translates into a 0.15% increase in corporate leverage, representing a $132 billion increase in aggregate leverage based on current values.

DOI:10.1561/104.00000069