ABSTRACT

Using firm-level panel data, this paper examines whether the cost of capital (COC) differs significantly between U.S.-based multinational corporations (MNCs) and U.S. domestic corporations (DCs). The results suggest that U.S.-based MNCs have higher COC than U.S. DCs and that industry importantly influences COC. The study also finds that there is a significant time effect on COC, and the time effect follows the trend of the U.S. economic growth rate. Using a Bayesian Markov chain Monte Carlo approach, we estimate jointly cost of equity, cost of debt, and capital structure, and find that the higher cost of capital for MNCs is due mainly to their higher cost of equity and greater use of equity financing; the cost of debt financing does not differ significantly for MNCs versus DCs.

Introduction

A large volume of research has explored differences between multinational corporations (MNCs) and domestic corporations (DCs) in the three components of overall cost of capital (COC): cost of debt financing (Mansi & Reeb, 2002; Reeb, Mansi, & Allee, 2001), cost of equity financing (Agmon & Lessard, 1977; Brewer, 1981; Fatemi, 1984; Forssbǽck & Oxelheim, 2011; Reeb, Kwok, & Baek, 1998), and capital structure (Burgman, 1996; Chen, Cheng, He, & Kim, 1997; Chkir & Cosset, 2001; Joliet & Muller, 2013; Lee & Kwok, 1988; Mansi & Reeb, 2002; Ramirez & Kwok, 2010). Since the results on each of the three components are mixed, and since overall COC depends on all three components, it is not possible to infer the difference between MNCs and DCs in COC. Little empirical work has compared COC for MNCs and DCs. This empirical study explores the effect of international diversification on corporate COC. This aspect of corporate activity is very important to understand the value impact on a firm of operating in the multinational domain. When firms select projects or when financial professionals evaluate a firm, the benchmark for return is usually the overall COC instead of cost of debt or cost of equity. The COC is also used to assess firms’ financial performance. For example, a widely accepted measure of financial performance, economic value added (EVA), is based on comparing the COC with the return on invested capital. This study will help in understanding how MNCs’ value is influenced by international diversification through its impact on COC. Previous studies on the relationship between international diversification and firm value have shown mixed results and have not focused on the mechanism through which the relationship occurred. Christophe (1997) and Denis, Denis, and Yost (2002) document a negative relationship between international diversification and firm value. Using data for the 1990s and early 2000s, Francis, Hasan, and Sun (2008) and Gande, Schenzler, and Senbet (2009) find that international diversification increases firm value, while Santos, Errunza, and Miller (2008) find that international diversification does not destroy value. This paper takes a further step by focusing on COC, one of the two factors that directly influence firm value, and helps in answering the question as to how international diversification influences MNCs’ value. This paper contributes to the literature in the following three ways. (1) Many studies on international diversification focus on whether international diversification affects firm value. This study elucidates how it does so, by using firm-level data to study the effect of international diversification on U.S. firms’ overall COC. (2) This study is the first to relate actual COC to real economic activities. (3) Using a Bayesian Markov chain Monte Carlo approach and the same set of data, we estimate jointly cost of equity, cost of debt, and capital structure to find out what causes the COC difference for MNCs and DCs. The three components of COC influence each other, but previous studies estimate only one of them at a time, a limitation that may render inconsistent estimation results. The rest of this paper is organized as follows. Section 2 reviews the literature. We present the data sources, sample selection, description of several variables, and descriptive analysis in section 3. We report the results of multivariate analysis for COC in section 4, and identify the origin of the COC difference between MNCs and DCs in section 5. Robustness checks are discussed in section 6. Section 7 concludes.