Lancelot Chardel*, Garance Desrousseaux**

*Using the ECB's AnaCredit supervisory register and Prentice--Williams--Peterson stratified Cox models, we study whether loan renegotiations between eurozone banks and non-bank financial intermediaries carry disciplinary content or mask forbearance. We document three findings. While, renegotiation occurrence and intensity carry no information about subsequent default ; the type of the opening amendment does. Favorable and unfavorable amendments both reduce default hazard, while mixed amendments sharply increase it. Group-affiliated NBFIs exhibit markedly lower baseline default hazard. Critically, the protective content of renegotiation is asymmetrically distorted for affiliates: favorable amendments lose informational content (dilution) while unfavorable amendments gain it (credibility through rarity). This asymmetric signature --- which no pure monitoring model generates --- is consistent with strategic forbearance inside banking groups and suggests that supervisors should track the type of renegotiations on bank–NBFI exposures, not only their frequency.

**Does relaxing financing constraints increase firm entry and firm performance? I exploit the introduction in 2006 of a French unemployment insurance reform, which allows unemployed individuals to convert part of their unemployment benefits into upfront startup capital. I implement a difference-in-differences strategy comparing sectors and municipalities differentially exposed to the reform before and after 2006. Using administrative firm-level data matched with unemployment insurance records, I examine the reform’s effect on both the quantity and quality of entrepreneurship, measured through firm survival, employment growth, indebtedness, and innovation-related outcomes.