The risk that action by the host country will reduce the value of the investment.

Formal and informal institutions are understood to influence firms' international investments, with host country regulatory environments playing a significant role in the attraction of foreign direct investment (FDI). Such “rules of the game”, set by governments (Pástor and Veronesi 2012), give rise to political risk, or uncertainty regarding a government's future actions (Pástor and Veronesi, 2012, Pástor and Veronesi, 2013), which represents an important source of risk and one expected to influence firms' FDI decisions (Stulz 2005; Giambona et al. 2017; Lin et al. 2019).1 Yet one that is difficult for managers and investors to assess with accuracy, as the opening quote illustrates, which could lead to capital misallocation (Bekaert et al. 2016; Col et al. 2018).

Dynamics between firm investment decisions and political risk are likely to be of particular significance in the context of FDI, given that multinational enterprises (MNEs) encounter considerable political uncertainty when choosing where to invest: balancing the benefits and costs offered by locations (Alcácer and Delgado 2016). Such trade-offs matter because benefits from investing in specific countries might be partially, or fully, offset by the risks of host governments taking arbitrary actions that may harm foreign firms (Desai et al. 2008), such as expropriation and property rights violations (Kesternich and Schnitzer 2010; Azzimonti 2018; Lin et al. 2019).

Although political risk is expected to influence MNEs' FDI decisions, exactly how and in what ways remains unclear. For instance, managers are unlikely to consider political risk independently of capital structure, given the potential for market discipline to influence investment (Harris and Raviv 1990; Lang et al. 1996; Aivazian et al. 2005; Ahn and Denis 2006). On one hand, a number of studies present evidence consistent with reduced investment when political risks are higher (Julio and Yook, 2012, Julio and Yook, 2016; Pástor and Veronesi, 2012, Pástor and Veronesi, 2013). For example, Pástor and Veronesi, (2012) provide theoretical predictions that political risk should reduce investment. Julio and Yook (2016) study cross-border investments, showing that political risk affects capital flows to foreign affiliates. On the other hand, although higher political risk should be associated with less investment on aggregate, the impact of political risk on investment can be heterogeneous across firms (Holburn and Zelner 2010; Pástor and Veronesi 2012; Jiménez et al. 2014). Firms may misallocate capital in the presence of foreign political risk (Bekaert et al. 2016; Col et al. 2018), and in some cases in order to offset political risks, firms might adopt low resource commitment entry mode (Slangen and Beugelsdijk 2010; Oetzel and Oh 2014). Finally, some firms may increase international investments when political risk is high (e.g., Holburn and Zelner 2010; Jiménez et al. 2014).

In this paper, we focus on decisions by MNEs as to how much to invest in FDI projects. We hypothesize that capital structure, and, more specifically, leverage, serves to condition the extent to which managers are willing to commit financial resources to FDI projects, and especially so in the presence of host country political risks. Our conjecture builds on the close relationship between firm leverage and political risk (Desai et al., 2004, Desai et al., 2008; Kesternich and Schnitzer 2010), and has roots in broader corporate finance and investment literatures, which establish that capital structure can influence investment in the presence of firm risks (Jensen 1986; Harris and Raviv 1990). This literature suggests that leverage plays an important role in disciplining agency costs of over/under investment (Lang et al. 1996; Aivazian et al. 2005; Ahn and Denis 2006).

We outline, in the Disciplinary effect, an important theoretical channel through which leverage influences FDI capital expenditure choices in the presence of political risk. For example, creditor monitoring could play an important role in disciplining investments decisions. Although leverage induces the possibility of firm failure, it may convey a disciplining effect on managerial behavior since the need to repay debt mitigates the agency costs of free cash flow (Jensen 1986; Harris and Raviv 1990). Such a disciplining effect could induce greater managerial effort (Jensen 1986) or, alternatively, sub-optimal investment (Col et al. 2018). The Disciplinary effect may be more pronounced in firms with lower growth opportunities, given that slow growing firms have limited investment options (Lang, 1996; Aivazian et al. 2005; Ahn et al., 2006).

We construct a project-level FDI dataset, sourced from FDI Markets: Cross Border Investment Monitor (a database provided by The Financial Times). The dataset contains valuable information at the project level, including identification of investing firms and their industries, locations of FDI, FDI type (category), and crucially, the capital expenditures allocated by investing firms to the FDI. Our dataset features outward foreign direct investment (OFDI) by MNEs originating from the United States (U.S.) to over 100 host economies with varying degree of institutional development and political risks. We supplement the FDI dataset with firm-level data from Compustat, and political risk data from the International Country Risk Guide (ICRG-PRS Group). Our final project-level dataset boasts 10,329 FDIs undertaken by 1135 MNEs from 2003 to 2014.

By capturing financial commitments using FDI project Capex (i.e. the level of capital expenditure per FDI project), and controlling for various firm-specific determinants of FDI and fixed effects (year, FDI project category, industry and country),2 our main result demonstrates that leverage and political risk3 are both associated with lower financial commitments. Furthermore, we present evidence of a dynamic interplay between leverage and host country political risk. When political risk is high leverage conveys an especially strong disciplining effect on firms' commitment of financial resources, whereas, as political risk reduces, the effect of leverage loses power and converges to zero at much lower political risk levels. In terms of economic significance, we estimate that when political risk is substantial (for instance, in the case of investments located in the BRIC countries - Brazil, Russia, India and China), a 10% increase in leverage is associated with a 1.7% reduction in FDI Capex. This effect gains significant economic relevance especially when the typical investment is large, such as for manufacturing and extraction FDI and in industries such as oil and gas, energy and automobiles. In sum, leverage constraints political risk exposure, with this effect growing in magnitude as political risk becomes more acute.

Next, we explore channels through which capital structure may influence firms' financial commitments to FDI projects. Since the Disciplinary effect may be more pronounced for low growth firms, we interact leverage with a low growth dummy (capturing firms with low Tobin's Q ratios), and find supportive evidence that leveraged firms are less willing to commit larger financial resources to FDI when growth opportunities are low. We view these results as theoretically consistent with the Disciplining Effect of leverage on over-investment, since leverage moderates risk-exposure, and especially so when host country political risk is high and growth opportunities low. We also interact leverage with interest coverage, to test whether the disciplining effect is stronger when managers are committed to paying higher interest on firm debt. Since risky debt disciplines investment, a riskier debt profile (higher cost) should intensify the disciplinary effect. We find evidence this is indeed the case. Finally, because larger firms may have greater debt capacity and are less resource constrained, we consider the effect of firm size and find that the disciplinary effect of leverage on FDI Capex is significantly weaker (stronger) for larger (smaller) firms.

We also consider several specific FDI project types (categories): manufacturing (characterized by higher sunk costs and lower flexibility); technological (R&D and internet) (which may be inherently riskier); and sales and retail (relatively more flexible given lower capital commitments). We find that the disciplining effect of leverage is stronger for technological and manufacturing FDIs but insignificant for sales and retail investments. Hence, when projects are riskier, and firms incur higher sunk costs and have less flexibility (higher exit barriers), leverage and political risk become increasingly important, with the disciplinary effect of leverage in the presence of political risk further increasing. Thus, the disciplining effect varies by project category and host market riskiness.

While our findings are robust to various controls and fixed effects, suggesting they are unlikely to be driven by omitted variable bias, endogeneity can arise in other ways and potentially affect our results. Causality issues may persist if firms adjust leverage as a response to political risk (Desai et al. 2008). To address endogeneity, we exploit plausibly exogenous variation in firm leverage arising from U.S. state-level unemployment insurance benefits (UI). Increases in UI affect firm leverage through their effect on workers' exposure to unemployment risk. By rendering lay-offs less costly, unemployment benefits reduce workers' compensation demands from firms in the event of job loss. Thus, firms located in U.S. states with more generous UI benefits choose significantly higher leverage (Agrawal and Matsa 2013), making state-level UI benefits a credible source of exogenous variation in firm leverage. Consistent with this, we show that increases in UI benefits are associated with exogenous increases in leverage, and that leverage instrumented through state-level UI benefits has a negative effect on FDI Capex which becomes significantly stronger when political risk is higher. These results provide evidence that our findings are robust to endogeneity concerns.

We also perform batteries of robustness checks. We substitute our main measure of foreign investment (project ln Capex) with a measure of the number of jobs created by each FDI project (ln Employment), ensuring that our findings are robust to alternative foreign investment measures. We control for several country specific observable characteristics, since although our models control for country fixed-effects, there may still be country observable effects that correlate with political risk and influence FDI Capex. Our findings remain robust after adding country controls.

We make several contributions. First, we advance understanding regarding the determinants of firms' FDI decisions. By focusing on the role played by capital structure in influencing MNEs' willingness to commit resources to FDI projects in the presence of host country political risks, we contribute to an unsettled debate regarding the role of leverage in influencing international investments. For example, it has been suggested that FDI may exacerbate agency costs of debt due to weaker monitoring (Doukas and Pantzalis 2003) and that firms may misallocate capital in the presence of foreign political risk (Col et al. 2018). We add to knowledge by identifying, in firm leverage, a novel mechanism conditioning the nature of FDI, which we demonstrate reduces the level of financial commitment to FDI projects.

Second, we contribute to literature analyzing the impact of leverage on firm investments (Lang et al. 1996; Aivazian, 2005; Ahn and Denis 2006). To the best of our knowledge, we are the first to demonstrate the relevance of leverage as a monitoring mechanism affecting the agency costs of over/under investment in the context of firms' FDI decisions, and also to explore such a rich, detailed and granular FDI database in corporate finance research. We find that the disciplining effect varies with host country political risk, firm growth opportunities, financial distress risk and size, and the riskiness of FDI project categories. Given that firm-level determinants of FDI are still little understood, our paper makes an important step in that direction, and adds to understanding on the impact of foreign political risk on investment (Desai et al. 2004; Desai et al. 2008; Kesternich and Schnitzer 2010; Pástor and Veronesi 2012; Julio and Yook, 2012, Julio and Yook, 2016; Azzimonti 2018; Col et al. 2018).

The rest of the paper is organized as follows. Section 2 reviews literature and formulates hypotheses. Section 3 outlines data, discusses variables and presents empirical methods. Section 4 presents the main results. Section 5 presents additional robustness tests. Section 6 concludes.

Leverage and investment

Tension exists in the relationship between leverage and firm investment. Such tensions attributable to agency costs may have significant implications for FDI and may act to reduce or increase the riskiness and levels of investments of leveraged firms. The debt as a Disciplinary effect channel suggests risky debt should exert a disciplining effect on firm management; serving to reduce agency problems and preventing overinvestment (Jensen 1986). One important source of such discipline relates to

Data and research design

The dataset employed in this study is constructed from project-specific FDI data sourced from FDI Markets: Cross border Investment Monitor (an online database provided by The Financial Times).4

Main results

Estimation results are reported in Table 3.

We begin by commenting on the model reported in column (1), which is an initial specification including firm leverage, country political risk and the interaction of these two variables. The effect of firm leverage on FDI Capex is statistically significant and negative. This result corroborates hypothesis H1. The effect of political risk is significantly positive. Since higher scores on the political risk scale reflect lower levels of political risk,

Labor employment as a proxy for investment

In this subsection, we run robustness checks by relaxing our measure of foreign investment. We replace ln Capex with ln Employment (the natural logarithm of the number of employment posts generated by the FDI). The, we re-estimate Eq. (1). Results are shown in Table 8.

As per the results reported in column (1), we find that leverage has a significantly negative effect on the FDI's employment. Firms employ more (less) workers in FDIs located in countries with lower (higher) political risk.

Conclusion

This paper questions whether firms' capital structure matters for FDI capital expenditures decisions into countries with varying degrees of political risk. To implement our analysis, we construct a novel dataset containing 10,000 unique OFDI projects matched with 1135 distinct U.S. firms over the period 2003–2014. Our main results demonstrate that corporate leverage influences firms' risky FDI capital allocation decisions. In particular, we offer novel evidence that firm leverage exerts a

Recommended articles (6)