**1. Introduction**

How to financially motivate executives has attracted a great deal of attention in literature. This is not surprising, as self-interested managers—along with their substantial influence on firm policies—are found to extract private benefits at the expense of shareholders, and therefore a well-designed compensation contract is critically important for shareholders who seek to maximize their value. However, the dominant theme of existing literature in executive compensation only focuses on traditional components, such as a base salary, bonus, and equity compensation (stocks or options), but overlooks another important piece, pension benefits, which mainly include retirement plans and deferred compensation. As emphasized by Jensen and Meckling (1976), the ingredients of an ideal compensation package for managers should include both equity-based and liability-based instruments. The binding claim in liability-like benefits is expected to influence managerial risk-preference, mitigate the agency cost, and hence affect possible value reallocation among stakeholders.

Compared to the widespread practice of pension plans or retirement benefits, studies regarding to the role of such liability-based compensation are generally scant. Edmans and Liu (2011) and Sundaram and Yermack (2007) are the earlier studies that explore liability-based instruments in compensation from theoretical and empirical perspectives. Particularly, Edmans and Liu (2011) propose that pension compensation outperforms cash compensation and helps solve the agency conflicts between shareholders and debtholders. Sundaram and Yermack (2007) document that liability-based compensation accounts for an important portion of executives' total wealth, motivating them to invest conservatively and therefore reducing default risk. Recent studies suggest that executives' pension benefits are associated with lower CDS spreads (Wei and Yermack 2011), lower R&D investment and financial leverage (Cassell et al. 2012), lower cost of private loan financing and fewer restrictive covenants (Anantharaman et al. 2014), cash holding (Liu et al. 2014), and managerial risk-taking (Anantharaman and Lee 2014).

The research on incentive design is important not only because it, per se, targets a fundamental business contract, but also yields instructive and practical insights into how an agent's risk attitude is financially effected and how the stakeholders are exposed to the related consequence. As noted above, there is a paucity of research on the nature of pension incentive and its impacts on the risk management of firms. For example, Anantharaman et al. (2014) found that firms with more pension incentive are charged with lower cost of debt financing and fewer restricting covenants. Cassell et al. (2012) and Liu et al. (2014) observe that the use of pension incentive is associated with lower R&D investment and more cash holding. However, none of them provide an explicit link between pension incentive and corporate risk management, particularly with the interplay of shareholder governance. By utilizing the data of currency hedging strategy used by the MNCs, this paper aims to investigate a direct impact of pension incentive on the managerial decisions. Additionally, the unique and hand-collected hedging data across the industries offer us a chance to have an unambiguous view of active risk management.

Corporate risk management, particularly hedging strategy, has become a critical dimension of financial policy and also attracted a lot of research interest. The study based on the Wharton survey of derivative usage shows that a growing number of CEOs view financial derivatives as indispensable tools in managing firms' risk (Géczy et al. 2007). The academic research also documents that hedging increases firm value by overcoming the market imperfections, such as deadweight costs related with bankruptcy risk, aggressive tax region, the underinvestment problem, and high cost of capital (Smith and Stulz 1985; Froot et al. 1993; Rogers 2002; Campello et al. 2011). More importantly, the literature suggests that executives' compensation incentives are an important determinant of corporate hedging. The traditional view of financial incentives focusing on equity compensation predicts that granting stocks or stock options will motivate managers to overcome risk aversion. In contrast, Sundaram and Yermack (2007) argue that executives' pension benefits would generate different incentives. The nature of unsecured and unfunded pension liabilities may expose managers to firm risk due to a lack of diversification and may result in a devastating loss of personal wealth in the case of firm bankruptcy. Since hedging reduces a firm's cash flow volatility and consequently lowers the likelihood of bankruptcy, it is plausible that CEOs who hold a larger amount of wealth in pension will be more actively involved in managing firm risk through hedging.

This study focuses on currency risk management by using foreign exchange derivatives, because Géczy et al. (1997) investigated the use of derivatives for a sample of Fortune 500 non-financial firms and found that currency derivatives are used most frequently by corporations. Foreign currency exposure is also considered as a major source of risk by the US firms (Bodnar et al. 1998; Krapl and White 2016). It is arguable that a firm may have little need to use foreign currency hedging if it has no relevant foreign currency risk. Therefore, in the spirit of the literature (Graham and Rogers 2002; Campello et al. 2011), we identify the multinational firms with ex ante exposure to foreign currency risk in this study. However, we also followed Doukas and Pantzalis (2003), to consider the case that a firm might be subject to foreign currency risk due to the competitive environment. To do so, we included keywords related to foreign currency and foreign exchange market risk in our textual searching program to read if firms explicitly state their foreign currency exposures in 10-K filings.

In this paper, we first examine the relationship between pension incentive and hedging propensity. Based on the sample retrieved from the COMPUSTA Segment database and matched with EXECUCOMP executive pension database, our investigation covers 1625 US firms from 2006 through 2015. Hedging data are collected from 10-K filings compiled in the SEC EDGAR database. We find that a higher level of pension incentive is associated with a higher probability of adopting currency hedging strategy. A one-standard-deviation increase in the pension benefits in dollar amount leads to an increase of 2.2% in hedging probability. Meanwhile, a one-standard-deviation increase in the

pension proportion in CEO total compensation increases hedging probability by 6.5%. In addition, by using the notional amount of hedging position, we find that a higher level of pension incentive also brings about a larger position of hedging. For one percent increase in CEO pension relative leverage, the hedging position rises by 40 basis points of total assets (0.4%), an equivalence of USD 35.95 million investment in foreign currency derivatives.

As with most studies in business research, the endogenously determined incentive might be a potential concern in our research. However, the significant and positive relation between hedging activity and pension incentive remains consistent when we perform robustness checks, instrumental variables (IV) model to control for endogeneity. The results from the robust models suggest that endogeneity cannot explain away the positive impact of pension incentive on hedging activity, and the inference from the baseline models is unaffected when correcting for endogeneity.

Given a positive relation between pension incentive and hedging activity, a natural question to further explore is whether this relation is contingent on governance mechanism. To answer this question, we propose the optimal contracting hypothesis. Our analysis suggests a more prominent influence of pension incentive on hedging for the firms with strong shareholder power than those with weak shareholder power. Furthermore, we employ the model of Faulkender and Wang (2006) to quantify the impact of hedging on firm value, augmented with pension incentive and governance mechanisms. The empirical evidence shows that one dollar of investment in existing hedging position generates USD 0.374 to shareholders and one dollar increase in hedging position creates an additional value of USD 0.204. More importantly, we find that the marginal impact of pension incentive on the value of hedging is higher for the firms with strong shareholder power, which supports the optimal contracting hypothesis.

In the extant research, the influence of pension incentive on active risk management (e.g., hedging) has not been given enough attention. Belkhir and Boubaker (2013) and Krapl and White (2016) are the two of few studies that have explored this area. Belkhir and Boubaker (2013) examined pension paid to CEOs in the US bank holding companies and find that higher CEO pension holdings relate to higher use of derivatives to hedge against banks' interest rate risk. They explain that pension compensation binds the banks' default risk with the executives' interest and curbs their excessive risk-taking activities. Krapl and White (2016) document a negative relation between foreign exchange exposure and pension-based compensation paid to executives including CEOs, CFOs, and other top managers. These results imply that pension compensation encourages executives to reduce cash flow volatility and hence lowers firms' exposure to foreign exchange risk.

Our findings are consistent with Belkhir and Boubaker (2013) and Krapl and White (2016), but this study differs from them. First, we build a large sample of companies that span across a wide array of industries, rather than only financial institutions. This large sample recognizes the possible heterogeneity and produces more applicable conclusions. Moreover, as suggested by Krapl and White (2016) foreign exchange exposure is a major source of risk to most of the US firms. Our focus on currency risk management reflects such an urgent demand in the age of globalization. In addition, our hedging data allow a direct look at the financial derivatives used in risk management. This study supplements the literature with the supportive evidence and further casts new light on the long-debated contracting theory of executive compensation.

In this paper, we fill the void in the literature by empirically examining the impact of pension incentive on firm risk management and how such influence is contingent on governance environment. Our study contributes to the literature in the following ways. First, we document a significantly positive impact of pension incentive on foreign currency hedging implemented by the multinational firms. Second, this paper complements the literature of compensation design by providing new evidence to support the optimal contracting hypothesis. Our analysis suggests that pension, the liability-based instrument adopted in the environment of strong shareholder power has a more pronounced impact on executives' hedging decisions. In addition, this study also enriches the existing hedging literature. After controlling for other well-documented determinants, we find pension incentive plays a remarkable

role, which is different from the equity-like compensation, in determining managerial risk attitude gauged through the hedging strategy. Finally, this research relates to the literature of risk management and we add the new insight to understand the active currency hedging strategy used by the multinational companies in creating shareholder value.

The remainder of this paper is structured as follows: Section 2 will review the literature and develop the testable hypotheses. Section 3 describes the sample, the variable constructions, and the summary statistics are also provided. Section 4 discusses the model specification and reports our main empirical results. In Section 5, we present an extended analysis on the value of hedging, interacted with pension incentive and governance. Section 6 concludes the paper.
