Why Do Small Businesses Take on High Levels of External Loans? A Censored Quantile Regression Analysis

Article excerpt

INTRODUCTION

Small businesses created 64 percent (14.5 million) of the net new jobs during the period of 1993-2008 (U.S. Small Business Administration (SBA), 2009). Though they make significant contributions to the economy, these firms are especially vulnerable to economic downturns. U.S. SBA reports that the number of small business bankruptcies increases from 19,695 in 2006 to 43,546 in 2008. Business failures may be largely influenced by the firms' decisions to take on debt obligations. Accordingly, we focus on the capital structure decisions of small businesses with high levels of external loans. This distinguishes our study from prior studies that commonly assume consistent impacts of capital structure factors throughout the conditional distribution of debt. Further, we utilize data from a time period that resembles the period preceding the recent economic collapse in terms of small businesses' debt usage and economic conditions. Clarifying the factors of small businesses' debt decisions may help improve their survival during economic downturns given that such decisions in preceding economic expansion can have a long-term impact on business performance.

Tax laws allow businesses to reduce their tax liability by deducting interest expense from taxable income. This lowers explicit costs of interest payments to creditors. The seminal work of Modigliani and Miller (1958; 1963) conjectures that the deductibility of interest expense makes debt a more attractive source of capital than equity. However, borrowings are associated with implicit costs. For example, bankruptcy can occur when a business is unable to meet required interest and/or principal payments. Small businesses often fail to fully factor in the implicit costs of debt during the periods of economic growth. This can be dangerous as these firms generally lack excess assets to survive declining sales revenue and/or higher interest rates during subsequent economic downturns.

Among the suggestions on how to prevent future financial crises, one recommendation from Minneapolis Fed President is to limit the deductibility of interest payments of debt (Reuters, 2011). Similarly, the International Monetary Fund (IMF) presents the argument against the tax incentives of debt financing "given the large potential macroeconomic damage from excess leverage" (IMF, 2009). Though the suggestions on lowering the tax benefits of debt are not specific to small businesses, these firms could be affected by the potential changes in tax regulations. It is therefore necessary to examine whether tax savings from deducting interest payments have a significant influence on small businesses' decisions to take on high levels of debt. Such analysis is made possible through using the Censored Quantile Regression (CQR) that reports the effect of explanatory variables at various points in the conditional distribution of the dependent variable. Specifically, our CQR results of the 90th quantile apply to the firms with external loans that are higher than 90% of other small businesses.

Closely-held small businesses can choose from multiple organizational forms; however, they can be broadly classified as either C-Corporations or flow-through entities based on the firms' tax status. C-Corporations pay income tax while flow-through entities do not. In this case, tax incentives for debt usage may be different between the two types of the entity forms. In addition, unlike C-Corporations, flow-through entities have limited access to equity capital and their owners may not have the protections for the liabilities. We thus study debt usage by different entity forms of small businesses.

We find that marginal tax rates and profitability are subject to the changes in their explanatory power for the debt usage as the level of external loans increases. Otherwise, our results for high levels of external loans are similar to those for medium levels of the loans in terms of the impact of firm characteristics, industry factors, and owner characteristics. …

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.