Contents lists available at ScienceDirect Journal of Business Research journal homepage: www.elsevier.com/locate/jbusres Earnings and capital management in European banks – Combining a multivariate regression with a qualitative comparative analysis ☆ Inês Pinto ⁎ , Winnie Ng Picoto 1 ISEG, Lisbon School of Economics & Management, Universidade de Lisboa, Portugal ARTICLE INFO Keywords: Earnings and capital management Sovereign debt crisis Financial assistance Banking industry Multivariate regression fsQCA ABSTRACT In this paper, we analyze the effects of the 2008 financial crisis and the ensuing sovereign debt crisis on the quality of financial reporting in European banks by investigating the existence of earnings and capital man- agement. The sample comprises countries for which the debt crisis was more severe (Greece, Portugal, Ireland, and Italy) and two major European economies (France and Germany) for which the crisis was not as severe. The data analysis consists of a multivariate regression to examine the correlation between operating income and banks' regulatory capital via loan loss provisions. Further, we use a fuzzy-set qualitative comparative analysis (fsQCA) that indicates causal paths for the loan loss provisions. The multivariate results indicate that bank managers use loan loss provisions to manage earnings and regulatory capital during the sample period. The findings do not provide clear evidence of a decrease in managerial discretion after the 2008 financial crisis. Nevertheless, in the severely affected countries, the results indicate that the level of earnings and capital management decreases. Further, fsQCA shows that loan loss provisions exist and that a bank's size and its nonperforming loans can play key roles in their existence. 1. Introduction The shock of the 2008 financial crisis and the ensuing sovereign debt crisis in Europe created a major public debate on the regulation and supervision of the financial system. This debate focuses on the transparency and quality of financial reporting. Several papers identify the use of loan loss provisions (LLPs) in the banking industry to manage earnings and regulatory capital ratios to reduce income volatility and to avoid the costs from violations of capital requirements (Anandarajan, Hasan, & McCarthy, 2007; Curcio & Hasan, 2015; Leventis, Dimitropoulos, & Anandarajan, 2011). While a consensus exists in the research that LLPs are a tool for managing earnings, the empirical results do not always support the capital management hypothesis (Leventis et al., 2011). The recent re- search finds that the effect of the 2008 financial crisis on the quality of financial reporting in general and earnings management in particular has decreased financial misrepresentation (Cimini, 2015; Filip & Raffournier, 2014). Nevertheless, the literature about the effect of the crisis on earnings and capital management in the European banking industry is still scarce and unclear. Our aim in this paper is twofold: First, we investigate the existence of earnings and capital management in European banks. Second, we analyze whether and how the financial crisis and the ensuing sovereign debt crisis affected this type of misrepresentation in financial reporting. This paper extends the literature as we investigate the existence of earnings and capital management in the banking industry after the 2008 financial crisis. According to Lobo (2017), the crisis provides an ideal setting to study bank managers' discretion regarding reporting choices for risk taking and financial stability. In this context, we analyze a sample of listed and non-listed European banks during the period from 2007 to 2014. The sample consists of banks from two groups of countries: Greece, Portugal, Ireland, and Italy; and France and Ger- many. This grouping gives a better understanding of the effects of the debt crisis on different countries. To accomplish the research objectives, we apply a multivariate re- gression to investigate the correlation between the banks' operating income and regulatory capital and their use of LLPs. Then, we apply the fuzzy-set qualitative comparative analysis (fsQCA) to uncover more causal paths behind the occurrence of LLPs. Because fsQCA considers all causal conditions together to explain the outcome of interest, several https://doi.org/10.1016/j.jbusres.2017.12.034 Received 18 June 2017; Received in revised form 15 December 2017; Accepted 18 December 2017 ☆ The authors are grateful to financial support from FCT-Fundação para a Ciência e Tecnologia (Portugal), national funding through research grant (UID/SOC/04521/2013) and to Marina Castilho for her assistance with data collection. The author gratefully acknowledges the helpful comments and suggestions provided by the editors, two anonymous reviewers, and theparticipants of Gika 2017 conference. ⁎ Corresponding author at: Rua Miguel Lupi, 20, 1249-078 Lisbon, Portugal. 1 Rua Miguel Lupi, 20, 1249-078 Lisbon, Portugal. E-mail addresses: inespinto@iseg.ulisboa.pt (I. Pinto), w.picoto@iseg.ulisboa.pt (W. Ng Picoto). Journal of Business Research xxx (xxxx) xxx–xxx 0148-2963/ © 2017 Elsevier Inc. All rights reserved. Please cite this article as: Pinto, I., Journal of Business Research (2017), https://doi.org/10.1016/j.jbusres.2017.12.034