This paper examines whether the risk of a future collateral fire sale affects lending decisions. We study US mortgage applications and exploit exogenous variation in foreclosure frictions for identification. We find lenders to be less likely to approve mortgages when anticipated losses due to uncoordinated collateral liquidations are high, and when there is elevated risk of joint collateral liquidation. These results suggest that fire sale risk has implications for credit allocation, and that lenders' collective origination decisions mitigate fire sale risk ex-post. However, we also find the effects to be significantly weaker outside periods in which fire sales are salient.

Conference presentations: EUR seminar series, PhD Finance forum - AEFIN, European Economic Association (EEA) conference 2020, 7th Emerging Scholars in Banking and Finance (2020), 3rd QMUL Economics-Finance Workshop, De Nederlandsche Bank, 2021 RiskLab/BoF/ESRB, EFiC 2021, The Finance Symposium 2021; Norges Bank-CEPR 2021, 5th Benelux Banking Research Day, FIRS 2022, EFA 2022;

Coverage: World Bank: All About Finance.

This paper shows that non-banks become more competitive lenders when they have access to information about a firm's cash flows. For identification, we exploit a government policy that forced nine banks in the UK to share data on their existing borrowers’ current accounts with other lenders. We find that eligible SMEs, i.e. those with turnover just below £25 million, are more likely to start a new relationship with non-banks post-reform compared to firms above the cutoff. This result holds despite the firms’ credit histories already being observable suggesting a unique role for information on cash flows. The policy has modest real effects, as affected firms increase their liquidity buffers rather than investment. This evidence suggests that open banking can alleviate the hold-up problem, and stimulate competition in the SME lending market.

[Draft coming soon] 

Abstract: We investigate the effectiveness of business shutdowns to contain the Covid-19 disease. In March 2020, Italy shut down operations in selected sectors of its economy. Using a difference-in-differences approach, we find that municipalities with higher exposure to closed sectors experienced subsequently lower mortality rates. The implied life savings exceed 9,400 people over a period of less than a month. We also find that business closures exhibited rapidly diminishing returns and had large effects outside the closed businesses themselves, including spillovers to other municipalities. Overall, the results suggest that business shutdowns are effective, but should be selectively implemented and centrally coordinated.

Coverage: BlueSky, BizEd AACSB International, Growth Hub, lavoce (in Italian).

Data: 10.25397/eur.14500491 and 10.25397/eur.14500449

Abstract: In statistics, samples are drawn from a population in a data-genetaring process (DGP). Standard errors measure the uncertainty in sample estimates of population parameters. In science, evidence is generated to test hypotheses in an evidence-generating process (EGP). We claim that EGP variation across researchers adds uncertainty: non-standard errors. Our crowd-sourced empirical analysis finds that non-standard errors are sizable, on par with standard errors. Their size (I) co-varies only weakly with team merits, reproducibility, or peer rating, (ii) declines significantly after peer-feedback, and (iii) is underestimated by participants. 

Selected conferences: Microstructure Exchange, FIRS, RBFC, SED, SoFiE*, Vieco, and WFA.

*Runner-up for the best paper prize.

Coverage: Die Zeit.