Giuditta Perinelli

We study how firms respond to temporary liquidity injections by leveraging a large-scale debt moratorium introduced in Austria in 2020. Debt moratoria are widely used during crises, requiring banks to temporarily suspend loan repayments at borrowers' request. Using a discontinuity in eligibility based on pre-policy size, we estimate a marginal propensity to invest of 76 cents per euro of postponed payments. While the moratorium increased investment and profitability without increasing defaults once repayments resume, we document an incomplete take-up rate of 44\%. We rationalize these findings in a dynamic investment model and compare moratoria with grants, an alternative form of business support during downturns. Unlike grants, we find that moratoria induce self-selection: financially constrained firms with profitable investment opportunities are more likely to opt in. A benefit-cost analysis shows that, for the same fiscal cost, both policies generate similar gains in firm value, but moratoria stimulate substantially more output.

We study how the source of inflation and the investor base jointly shape portfolio rebalancing. We distinguish between cost-push ("bad") inflation and demand-driven ("good") inflation. Using expectations from the Survey of Professional Forecasters as a proxy for sophisticated investors' beliefs, we show that bad inflation shocks induce stagflationary expectations, whereas good inflation shocks raise expected inflation and improve expected economic activity. Remarkably, household expectations move in a similar way. Using security-level holdings of bond and equity mutual fund shares, we show that portfolio rebalancing is broadly consistent with these shifts in expectations: bad inflation generates outflows from both bond and equity funds, whereas good inflation generates outflows from bond funds and inflows into equity funds. Our identification strategy compares flows across fund shares held by different investor types within the same fund, holding the underlying portfolio risk fixed. We show that the investor base determines how rebalancing unfolds: investment funds rebalance rapidly and strongly, households adjust more gradually but persistently, and insurance companies rebalance more selectively. Within bond funds, portfolios with higher residual maturity-our proxy for duration exposure-experience larger return losses following both types of inflation shocks. Residual maturity matters particularly for institutional investors: following good inflation, flows by investment funds and insurance companies are consistent with a shift from higher-to lower-maturity funds, whereas household flows vary less with maturity. Following bad inflation, investors redeem from both maturity segments. The source of inflation therefore determines the direction of portfolio rebalancing while investor composition and duration exposure determine its speed, persistence, and distribution across asset classes and maturities.

We develop a model of credit cycles in which specialized lenders can preserve borrower going-concern value after foreclosure, allowing firms to borrow against value that arm’s-length financiers cannot fully realize. Limited pledgeability arises because this recovery advantage cannot be fully transferred to the intermediary’s own creditors, so specialized lending requires intermediary balance-sheet resources. With heterogeneous borrowers, specialized intermediaries expand credit especially to firms with less transferable, more intangible value. Following sufficiently large adverse shocks, however, scarce intermediary resources induce a flight-to-pledgeability, shifting new lending toward borrowers with higher recovery values. We then apply the model to private credit market and show that the presence of specialized intermediaries can explain several facts about lending cycles and private credit specialization. We also test and find supportive evidence for flight-to-pledgeability, using Business Development Company lending during COVID-19 as our laboratory. 

Does the investment response to inflation expectations grow with the forecast horizon? Existing evidence stops at one year. Using a randomized information experiment linked to balance sheet data on Italian firms, we show that the causal investment semi-elasticity nearly doubles from six months to three-to-five years, with the response four times larger for tangible than intangible assets. These patterns, with heterogeneity by leverage and debt maturity, point to a Mundell-Tobin channel: expected inflation erodes the real burden of fixed-rate debt, lowering the cost of new capital. A sufficient statistic mapping debt structure to the investment response formalizes the mechanism.