Firm-bank relationships - a cross-country comparison
B Size class classification
We use Bureau van Dijk’s Orbis to incorporate external firm-size characteristics which we use for firm size classification. To ensure the data quality we apply the following filters. First, we retrieve information only for EA countries. Second, we extract the unconsolidated codes (U1 and U2), as well as the code for firms which have only consolidated reports (C1), but no dependent entities.[18]Third, we retrieve files up to June 2019 and choose the nearest date for which we have 12-month financial statements. Finally, due to numerous duplicate files for each firm, we apply distinct logic to several key characteristics of the financial statements:
•Filing type - we prefer ‘Annual report’ over ‘Local filing’, if both are available
•Consolidation code - we apply in order of preference the codes U1, U2 and C1, with U1 being the first choice if available.
•Audit status - we prefer audited files if available.[19]
•Accounting practice - we prefer the ‘Local GAAP’ over ‘IFRS9’ and missing values.
To utilize all available information on balance sheet, number of employees, and annual turnover, we apply the following logic:
1.If all three firm characteristics are available or only the number of employees and oneof the remaining two, then we apply the official definition of the European Union (as described above).
2.If only the number of employees is not available, then we apply the official definition ofthe European Union (as described above) only to the balance sheet size or the annual turnover, depending on which has the higher value.
3.If only one of the three characteristics is available, then we apply the official definition of the European Union (as described above) only to that characteristic. For instance, if we have only the number of employees available and it has a value of 150, then the firm will be classified as medium sized.
By utilizing both data sources and applying the official definition of the European Union for the size classes we manage to classify 85.0% of the firms in AnaCredit in December 2019.
Acknowledgements
We thank seminar participants at the ASSA/AEA 2022 and the ECB for comments and discussions. Fabiano Schivardi thanks the ERC for financial support (ERC grant 835201). All errors and omissions are our own.
The views expressed in this paper are the responsibility of the authors only and should not be interpreted as reflecting the views of, or imply any responsibility for, the European Central Bank and the Eurosystem.
Kamelia Kosekova
European Central Bank, Frankfurt am Main, Germany; email: k.kosekova@gmail.com
Angela Maddaloni
European Central Bank, Frankfurt am Main, Germany; email: angela.maddaloni@ecb.europa.eu
Melina Papoutsi (corresponding author)
European Central Bank, Frankfurt am Main, Germany; email: melina.papoutsi@ecb.europa.eu
Fabiano Schivardi
Luiss University, Rome, Italy; Einaudi Institute for Economics and Finance, Rome, Italy; Centre for Economic Policy Research, London, United Kingdom; email: fschivardi@luiss.it
Postal address 60640 Frankfurt am Main, Germany
Telephone +49 69 1344 0
Website www.ecb.europa.eu
All rights reserved. Any reproduction, publication and reprint in the form of a different publication, whether printed or produced electronically, in whole or in part, is permitted only with the explicit written authorisation of the ECB or the authors.
This paper can be downloaded without charge from www.ecb.europa.eu, from the Social Science Research Network electronic libraryor from RePEc: Research Papers in Economics. Information on all of the papers published in the ECB Working Paper Series can be found on the ECB’s website.
PDF ISBN 978-92-899-6111-0 ISSN 1725-2806 doi:10.2866/505807 QB-AR-23-063-EN-N
[1] For an overview of the recent developments in the financial and corporate sector in the euro area see ECB (2021).
[2] We are aware of only one paper that carries out a similar analysis in intent, the work by Ongena & Smith (2000), which is based on evidence from a survey conducted in 1996 on 1,079 firms across 20 European countries.
[3] For further information on the AnaCredit project, see https://www.ecb.europa.eu/stats/money_ credit_banking/anacredit/html/index.en.html.
[4] AnaCredit has all information on bank loans starting from September 2018. We do not take the most recent data, because we don’t want our analysis to be confounded by the potentially significant effects of the Covid-19 pandemic.
[5] See Degryse, Kim & Ongena (2009) for a survey and Kysucky & Norden (2016) for a meta-analysis of the effects of relationship lending on access to credit in general. Beck, Degryse, De Haas & Van Horen (2018) show that relationship lending is particularly important during recessions, when firms need credit the most.
[6] Non-revolving credit lines are similar to loans, with the difference that the funds are given in preestablished tranches over time, rather than all at once.
[7] Consistent with our result, Feraboli, Häkkinen & Gutiérrez (2015) show that firms in countries more affected by the Great Financial Crisis have a higher share of short-term credit compared to firms in countries that were less affected.
[8] In this paper, we do not investigate potential explanations of the cross-country differences on the interest rates, as the main focus is to document those. Differences on the interest rates across countries are described also by the OECD. (2022) report, where issues with access to credit is highlighted as a key explanatory factor. Feraboli et al. (2015) show that the monetary policy pass-through during the low interest rates period was much more sluggish for countries more affected by the financial crisis and thus led to higher interest rates for firms in these countries compared to firms in countries in the euro area less affected by the financial crisis.
[9] AnaCredit reports all loans granted by credit institutions residing in euro area Member States, including loans granted by their domestic / foreign branches. The data is maintained by the ECB and the Eurosystem National Central Banks and is harmonized across Member States. A credit instrument is subject to reporting
[10] This threshold ensures that clearly implausible data points are discarded without significantly restricting the data coverage.
[12] Appendix Tables A1-A11 report the same information separately by country.
[13] For reference, the monetary policy rates in the euro area at the end of 2019 were -0.5 for the deposit rate, 0 for the MRO and 0.25 for the MLR.
[14] In fact, the regression drops one size and one sector dummy, so that the country dummy refers to that size-sector cell. Of course, the choice of the excluded sector and size class affects the absolute values of the ?c but not the cross country differences ?c1 -?c2. We re-base the coefficients by subtracting the average of the conditional country dummies 1/nPc?s and adding the average unconditional country means, so that by construction the overall country mean is equal for the conditional and unconditional means.
[15] To better summarize the information, we only report country aggregates in the main text. Appendix Figure A4 shows the share by firm size. The country rankings by instrument type tend to be less stable across size classes compared to the other variables that we consider in the analysis.
[16] Note that, compared to Equation 4 where the unit of observation is a firm-period, here we are at the more granular level of firm-instrument-period.
[17] In fact, if we run the regression with only country dummies, we recover exactly the unconditional mean rates.
[18] According to the Orbis Data Guide, companies with detailed financial data can have 4 consolidation codes - U1 (statement not integrating the statements of the possible controlled subsidiaries or branches of the concerned Company with no consolidated companion), U2 (statement not integrating the statements of the possible controlled subsidiaries or branches of the concerned Company with an consolidated companion), C1 (statement of a mother Company integrating the statements of its controlled subsidiaries or branches with no unconsolidated companion), C2 (statement of a mother Company integrating the statements of its controlled subsidiaries or branches with an unconsolidated companion)
