Debt Maturity and Tax Avoidance

Abstract

This study proposes and empirically tests the argument that creditors are likely to extend debt with a shorter maturity to tax-avoiding firms so that they can frequently re-evaluate tax-related risk in debt contracting. Using effective tax rates and uncertain tax benefits as a proxy for tax avoidance, I find that tax-avoiding firms have a larger proportion of short-maturity debt compared to other firms. The empirical findings further show that firms with unsustainable tax positions and with subsidiaries in tax-haven countries are more likely to employ short-maturity debt. Collectively, the empirical findings suggest that frequent debt renegotiations increase the exposure of tax-avoiding firms to credit supply shocks, contributing to their higher demand for cash.

Acknowledgements

This work has significantly benefited from the comments and recommendations of the editor (Laurence van Lent), Martin Jacob, Javier Gomez, anonymous referees, seminar participants at ESADE, conference participants at the Annual Meeting of the European Accounting Association 2012 and Second Workshop on Current Research in Taxation 2012.

Supplemental data and research materials

Supplemental data for this article can be accessed at 10.1080/09638180.2015.1106329.

Notes

¹Following Hanlon and Heitzman (2010), this paper views tax avoidance as encompassing a spectrum of tax-planning activities with outcomes ranging from certain to uncertain, where uncertain (i.e. aggressive or risky) tax positions are supported by a relatively weak set of facts and therefore less likely to be sustained upon a tax audit.

²Consistent with previous work, I refer throughout this paper to debt that matures in less than three years as ‘short-term debt’ and to debt that matures in more than three years as ‘long-term debt’.

³An illustrative example of this notion is the recently leaked documents regarding special tax rulings in Luxembourg. The International Consortium of Investigative Journalists and the Center for Public Integrity published 548 comfort letters issued from 2002 to 2010. These letters expose the deliberate efforts of large firms to reduce their taxable income by creating a complex organization via dubious transactions with related parties. Source: http://www.icij.org/project/luxembourg-leaks/leaked-documents-expose-global-companies­secret-tax-deals-luxembourg.

⁴The ‘under-sheltering puzzle’, highlighted by Weisbach (2002), suggests that firms have traditionally failed to use tax shelters to any significant degree.

⁵Christensen, Nikolaev, and Wittenberg-Moerman (2015) provide a comprehensive review of the existing theory and empirical work on the efficiency role of accounting information in debt contracting.

⁶Mills et al. (1998) estimates that an additional $1 investment in tax planning results in a $4 reduction in tax liabilities. More recently, Wilson (2009) suggests that sheltering activities can explain a reduction in effective tax rates of approximately 5.1% and (median) federal tax savings of $66.5 million.

⁷FIN 48 requires firms to evaluate their tax position by using a two-step process. First, a firm should recognize the financial statement benefit of a tax position only after determining that the relevant tax authority will likely sustain the position after an audit. Second, the amount recognized should be the largest benefit that has a likelihood greater than 50% of being realized upon settlement with the relevant tax authority. The amount not recognized is called the unrecognized tax benefit and is recorded as a tax-contingent liability. FIN 48 also contains important disclosure requirements, including a tabular reconciliation of the beginning and ending balances of unrecognized tax benefits (Frischmann, Shevlin, & Wilson, 2008).

⁸Scott Dyreng extracted the subsidiary locations from Exhibit 21 of 10-K filings and has generously provided access to the data set (Retrieved October 2014 from https://sites.google.com/site/scottdyreng/).

⁹The findings are robust to the estimation procedure. I obtain equivalent results using a Tobit regression model.

¹⁰The percentage change is calculated by multiplying the standard deviation and the regression coefficient: 0.104 × 0.17 = 0.018 for effective tax rates, and 0.027 × 0.19 = 0.005 for cash-effective tax rates.

¹¹Calculated as follows: −0.326 × 0.03 = −0.001.

¹²A rational manager prefers long-maturity debt, because it locks creditors into low rates, whereas future bad news is not factored into debt contracting.

¹³The effect is calculated as follows: (−0.046 × 0.17)/0.092 = 0.085, where the denominator is the average CFVOL calculated over following three years.

¹⁴The effect is calculated as follows: (0.108 × 0.17)/0.0178 = 1.033.

¹⁵Kubick, Lynch, Mayberry, and Omer (2014b) suggest that tax-avoiding firms are more likely to receive a tax-related US Securities and Exchange Commission (SEC) comment letter. An Online Supplementary Material presents additional empirical evidence using the issuance of tax-related SEC letters as an alternative proxy for tax avoidance. The empirical tests suggest that there is a decrease of 3.6% in firms’ debt maturity following the receipt of a tax-related comment letter.

¹⁶This effect is calculated as follows: −0.210 + 10×(−0.004) = −0.25.

¹⁷Recently, Cassell, Huang, Sanchez, and Stuart (2012) and Kubick, Lockhart, and Robinson (2014a) employed median industry values as an instrument in two-stage least-square regression models.

¹⁸The effect is calculated as follows: 0.872 × 0.17 = 0.1482.

¹⁹An Online Supplementary Material presents additional empirical evidence regarding the role of tax audits on debt maturity.

²⁰The simultaneous equations model satisfies the order and rank restrictions for identification. First, each equation includes at least one exogenous variable, thus satisfying the order condition for identification. Second, most instruments in the SEM are statistically significant at better than 1% significance level. Therefore, the equations in the SEM also satisfy the rank condition for identification.

Additional information

Funding

Petya Platikanova gratefully acknowledges financial support from Government of Catalonia [grant 2014-SGR-1079], Banc Sabadell and ‘la Caixa’.