Real and financial frictions and the dynamic properties of physical investment

This dissertation studies the dynamic properties of physical investment in a general equilibrium economy under various real and financial frictions. The first chapter studies plant-level irreversible investment in general equilibrium. I show that in contrast to previous studies, the irreversibility model fails to generate a smoother aggregate dynamics than a neoclassical benchmark and this conclusion is robust with respect to the level of uncertainty as long as the uncertainty level is time-invariant. This is because the consumption smoothing needs in general equilibrium destroy the value of waiting associated with the irreversibility. However, I find that when the uncertainty level itself fluctuates in the short run, the model generates a stronger incentive for waiting, thereby substantially modifying the basic properties of the neoclassical investment model.;The second chapter studies the role of financial market imperfection in the sudden collapse of aggregate investment spending during the Asian financial crises (1997). We approach the problem in two ways: a conventional reduced-form analysis of a panel data of Korean manufacturing firms and an indirect inference to estimate a structural dynamic programming problem of a firm with foreign debts and financial constraints. Both reduced-form evidence and structural parameter estimates imply an important role for finance in investment at the firm level. Counterfactual simulations imply that the effect of foreign denominated debt for investment spending may account for up to 50% of the drop in investment during the crisis period.;The third chapter studies the role of sunk initial investment required for export activity in generating endogenously persistent dynamics for trade and real exchange rate. With the sunk entry cost, exporting firms tend to stay in the business even when they make negative operating profit. On the other hand, non-exporters tend to delay entry decision. I find that the trade model based on sunk entry cost has a great potential to generate a more persistent dynamics of trade and real exchange rate. The key mechanism delivering this result is that the marginal cost of increasing the number of exporters is increasing because the productivity of marginal exporters is deteriorating as the economy continues to expand export activity.