Investment and liquidity management are analyzed in a sector where empirical estimates of Tobin’s q provide reliable signals of investment opportunity and where firms are exogenously cash constrained (these firms are known as real estate investment trusts, or REITs, which hold ownership positions in commercial real estate and are required to pay out at least 90 percent of income as dividends). Empirical modeling utilizes a 3SLS specification to account for the endogeneity of liquidity management and investment, where liquidity management occurs through the use of dividend (cash retention) policy and bank line of credit (L/C) usage.

Across the entire sector, we document substantial realized investment as well as high investment sensitivity to q. Investment is also sensitive to measures of financial market frictions, suggesting that constraints on retaining cash flow distort investment decisions. Liquidity is actively managed through dividend policy and access to short-term bank finance, in which bank L/C smoothes variation in available cash flow and accelerates investment. Bank L/C thus plays an important role in investment through both capacity creation and incremental usage, suggesting that debt overhang and preemptive investment are relevant financial market frictions. Bank L/C use is also highly responsive to investment, and realized cash flow is a direct substitute for L/C use.

Using the Kaplan-Zingales (1997) method for measuring the degree of financial constraint, we identify substantial differences between investment and liquidity management policies of firms, where more (less) financially constrained firms in our sample exhibit high (low) investment and liquidity management sensitivity to variables that measure financial market frictions. Altogether, our findings show that firms which are less financially constrained as measured by KZ index score are highly responsive to investment signals contained in their stock prices, but are mostly unresponsive to other variables that proxy for financial market frictions—including cash flow. There does appear to be interdependence between L/C usage and investment with the less constrained firms, but the effects are much less pronounced than with the more constrained firms. In comparison, the more constrained firms are unresponsive to investment signals contained in their stock prices, but display extreme sensitivity in all three structural equations to a number of variables that proxy for financial market frictions. Results suggest that firms that are cash constrained (which are all firms in our sample, due to dividend payout requirements) are not necessarily financially constrained.