This paper asks how changes in investors' risk-taking capacity affect asset prices and economic growth. I develop a Schumpeterian growth model with risk-averse investors. Risk limits raise the market risk premium, reduce R&D, and can dampen entry innovation in general equilibrium when the rise in the risk premium is sufficiently strong. Using a tightening of risk requirements on British pension funds as a natural experiment, I show that their subsequent divestment from equities led to a persistent fall in stock prices among firms more exposed to pension investors pre-reform. In response, these firms cut R&D expenditure and reduced long-term investment. Counterfactual simulations using a calibrated version of my model suggest that the impact of risk limits on growth can be sizeable and persistent when equity markets are inelastic.

We develop a model of endogenous growth and firm dynamics with soft budget constraints, where firms differ in their innovation speed and slower firms need additional financing in order to eventually innovate. As creditors cannot anticipate refinancing needs in advance nor credibly commit to withholding future refinancing, a Soft Budget Constraint Syndrome emerges, causing more activity by slow incumbents and crowding out potentially more efficient innovators. The resulting trade-off between the positive effects of budget constraint softening on innovation by incumbents and its negative effect on entry by fast innovators, generates a hump-shaped relationship between refinancing costs and aggregate growth. Calibrating the model to French firm-level data, we show that the budget constraint softening associated with the combination of an under-capitalized banking system and a decline in interest rates in the aftermath of the Global Financial Crisis accounts for 54% of the observed drop in the aggregate growth rates post-crisis. Although the softening in budget constraints has had a positive effect on incumbent innovation, this was more than offset by the resulting decrease in the entry rates of good firms.

This paper examines how bank-driven credit reallocation affects firm dynamics and economic growth. We develop a model in which firm creation and destruction are determined by banks' decisions to roll over or liquidate loans according to expected firm performance. We show, both analytically and quantitatively, that policies encouraging loan liquidation, such as insolvency law reforms, may reduce output in the short run by accelerating firm exit, but also promote firm entry and increase growth in the long run. Improvements at the exit margin can complement policies aimed at stimulating firm creation, such as R&D subsidies. This complementarity emerges because loan liquidation releases funds for new lending, making banks less reliant on external funding and lowering the equilibrium interest rate. If bank funding is inelastic, tighter capital requirements on banks can even increase lending to new firms, raising firm creation and growth.

We analyze how the government can promote innovation at different stages of firm growth and how optimal innovation policy interacts with private venture capital. Using a rich model of innovation financing and firm dynamics, in which only the most promising start-ups are financed and supported by venture capitalists, we demonstrate that  innovation policy optimally uses differentiated R&D subsidies to internalize spillovers, a profit tax to extract monopolistic rents, and invests in public basic research. Implementing optimal policy from scratch results in a long-run GDP gain of up to 12 percent, shifts start-up R&D to high-tech firms, and more than doubles the GDP share of venture capital. Welfare gains are equivalent to a permanent income flow of 0.8 percent of initial GDP.

I study the interaction between real interest rates and growth in an economy where competitive financiers have limited balance sheet capacity. Initial investments are made under asymmetric information, but after types are revealed, financiers can reallocate funds between projects. The optimal reallocation rate depends on the opportunity cost of reallocation, i.e. the real interest rate. As financiers do not internalise the feedback between real rates and growth, the equilibrium reallocation rate can be suboptimal: When real rates are low (high), there is too little (too much) reallocation. Because falling rates slow down reallocation, structural downward pressure on rates, e.g. due to demographics, has a hump-shaped effect on growth. With structurally low real rates, policies that encourage reallocation, such as a tax deduction on unrealised losses in the financial sector, can boost growth and move the economy closer to the growth-neutral real rate r**.