Website CV Paper
Fields
Innovation, Public, Macroeconomics
Job market paper
Experimental and Applied R&D Investment under Liquidity Constraints
When R&D support is unconditional on the type of research, does it matter whether it arrives as a subsidy or a refundable tax credit? We show it does. A subsidy lowers R&D's upfront cost, whereas a tax credit is paid only after expenses are incurred and, once received, becomes divertible cash the lender must cover. We embed limited-enforcement financial frictions in an endogenous growth model with heterogeneous firms conducting applied R&D, which raises current profit, and experimental R&D, which raises future productivity and, unlike applied R&D, spills over to other firms. Because the tax credit tightens the financing constraint, firms cut applied R&D, whose near-term profits are divertible, and reallocate toward experimental R&D, whose distant return they cannot pledge. Calibrated to Canadian data, this reallocation is concentrated among financially constrained innovating firms and, once aggregated across the firm distribution, the tax credit and the subsidy raise the mass of innovating firms by nearly the same amount, and the tax credit delivers 0.02 consumption-equivalent points more welfare than an equal-cost subsidy. Firm-level evidence is consistent with this mechanism.