Failure to correct for pension risk leads to upward-biased discount rate estimates in firms with pension risk exposure. The result is a negative and economically significant relation between pension risk and corporate investment. The effect is confined to investment decisions that require discount rate estimates. Moreover, it is stronger if project value is more sensitive to such estimates. Because of this bias, firms miss valuable investment opportunities. The results survive robustness tests that address endogeneity concerns and alternative interpretations of the evidence. The general implication is that non-operating risks can distort, if ignored, corporate investment decisions.

Previous studies show that economic policy uncertainty has been rising steadily since the 1960s (Baker et al. 2014), and that this secular increase has led to harmful economic outcomes such as reduced investment rates (Gulen and Ion 2016). Other studies find that politically connected directors play a unique role in corporate decision making and firm valuation (Goldman, Rocholl, and So 2009). We combine these two lines of research to examine the ability of politically experienced directors (PEDs) to mitigate the harmful investment effects of policy uncertainty. Our results show that the presence of at least one PED on a company board reduces the sensitivity of investment decisions to policy uncertainty by 49% relative to firms without PEDs. These results are stronger when the PED serves on a presidential (as opposed to legislative) committee, and among firms most vulnerable to investment irreversibility. We explore omitted variables bias and instrumental variable estimation in robustness testing to alleviate endogeneity concerns. Overall, our findings align with the theoretical framework of Pastor and Veronesi (2013) and real options theory, and confirm that PEDs can significantly reduce the harmful effects of policy uncertainty.