We use the new method to 21 countries with varying membership off economic and you will economic advancement. That it not simply adds to the external authenticity regarding prior to efficiency and in addition lets us mine ranging from-nation (in addition to within-country) adaptation in both the latest prevalence out-of relationships credit together with strength of one’s 2008–2009 economic downturn. Additionally, by using corporation-height questionnaire study, we can separate between economically unconstrained companies, firms that was basically possibly refused otherwise discouraged out-of making an application for a beneficial mortgage, and you can agencies as opposed to borrowing request. Training having fun with borrowing from the bank registry research deal with the possibility restriction one to non-candidate businesses possibly do not need credit, otherwise they require borrowing however they are disappointed out-of implementing. In the long run, because of the looking at the character out of relationships financing from the height and you can trough of your own borrowing from the bank period, we can reveal that, due to the fact forecast of the theory, relationships lending is far more extremely important through the a beneficial downturn. Regardless of it completely different empirical approach, our abilities line up well, which harden, before conclusions into character from dating lending throughout the downturns.
Three methodological factors need remark. Although not, that with studies both for borrowing from the bank and you can non-credit enterprises we could gauge the regional general equilibrium effect of banks’ lending processes, which we could possibly skip if attending to just into credit businesses. Second, i trust survey studies for both companies and you will finance companies. Whenever you are company-peak survey data had been popular regarding current literature (e.g., Campello ainsi que al., 2010, Popov and Udell, 2012, Ongena et al., 2013), concerns you are going to occur regarding the aspect mistake in the firms’ answers so you can issues on the as to the reasons they don’t really submit an application for a loan, otherwise you to definitely rejection you can expect to merely mirror having less capital ventures which have confident internet establish really worth. To mitigate such concerns, you can expect multiple robustness examination with assorted significance out-of firms’ borrowing limits, along with actions based on balance sheet data.
3rd, the empirical strategy hinges on the region away from banking companies and you will people are separate of every most other. Pursuing the Berger, Miller, Petersen, Ra), i believe that new financial land close providers imposes a keen exogenous geographical limitation on banks one to agencies get access to. We offer large research that’s consistent with this assumption. We together with decide to try having heterogeneous aftereffects of your regional visibility dil mil visitors away from banking companies one to examine by themselves because relationships lenders round the different kinds of enterprises. That it further contact the newest you can endogenous matching out-of providers and banking companies.
The fresh paper proceeds as follows. Part dos makes reference to the data supplies i blend, and you will Section step three gift suggestions the empirical method. We discuss our very own standard show and you may robustness examination when you look at the Sections 4 and you may 5, correspondingly. Part six next considers genuine consequences, and you will Area 7 stops.
Our very own empirical investigation rests towards the signing up for three essential pieces of suggestions: analysis on firms’ borrowing from the bank constraints in the different occuring times, the fresh new geo-coordinates of one’s financial twigs close these businesses, and you will studies for the lending process of those banking institutions. I talk about the data toward firms’ real abilities in the Part 6.
To estimate the link between the share of relationship banks near a firm and the probability that the firm is credit-constrained, we use the following model for both the 2005 and 2008–2009 cross section. We hypothesize that relationship banks were particularly helpful once the cycle had turned in 2008. Consider the model Y i j k l = ? 1 X i j k l + ? 2 L j k + ? 3 S h a r e r e l a t i o n s h i p b a n k s j k + ? 4 C k + ? 5 I l + ? i j k l , where Yijkl is a dummy variable equal to one if firm i in locality j of country k in industry l is