|
|
To do so, in this paper authors build a unique dataset where information on loan applications from SAFE is matched with firms’ and banks’ balance sheet information. Specifically, they combine the ECB proprietary SAFE-Amadeus-Orbis dataset, which augments SAFE with information on firms’ financial statements from Bureauvan Dijk’s Orbis and Amadeus – with data on banks’ asset quality, capitalisation, and profitability obtained from Fitch Connect. Using this comprehensive dataset, they are able to study whether the stronger financing obstacles faced by firms located in the periphery have been due to being less creditworthy than their peers in core countries, or rather due to confronting banks with a lower ability to grant credit.
Authors find that, while firm characteristics, and leverage in particular, are strongly associated with the probability of experiencing credit rejections, they leave a relevant part of the observed difference across periphery and core countries unexplained, even taking into account the incidence of firm distress in the periphery. On the other hand, their findings suggest that periphery-specific bank weakness is able to explain at least part of this difference. Authors also find evidence of a non-linear effect of the bank NPL ratio. Only beyond a certain level, a higher NPL ratio signals weakness in the bank balance sheet; at such higher levels, they find that it implies a lower ability to grant loans even to sound firms. Finally, authors do not find any evidence of capital mis-allocation, including in the euro area periphery.
They show that their results are robust to a number of different model specifications, definitions of credit rejections, as well as firm-bank matching criteria.