Springer Proceedings in Business and Economics Krzysztof Jajuga Lucjan T. Orlowski Karsten Staehr Editors Contemporary Trends and Challenges in Finance Springer Proceedings in Business and Economics Springer Proceedings in Business and Economics brings the most current research presented at conferences and workshops to a global readership The series features volumes (in electronic and print formats) of selected contributions from conferences in all areas of economics, business, management, and finance In addition to an overall evaluation by the publisher of the topical interest, scientific quality, and timeliness of each volume, each contribution is refereed to standards comparable to those of leading journals, resulting in authoritative contributions to the respective fields Springer’s production and distribution infrastructure ensures rapid publication and wide circulation of the latest developments in the most compelling and promising areas of research today The editorial development of volumes may be managed using Springer’s innovative Online Conference Service (OCS), a proven online manuscript management and review system This system is designed to ensure an efficient timeline for your publication, making Springer Proceedings in Business and Economics the premier series to publish your workshop or conference volume More information about this series at http://www.springer.com/series/11960 Krzysztof Jajuga • Lucjan T Orlowski • Karsten Staehr Editors Contemporary Trends and Challenges in Finance Proceedings from the 2nd Wroclaw International Conference in Finance Editors Krzysztof Jajuga Finance Management Institute Wrocław University of Economics Wrocław, Poland Lucjan T Orlowski John F Welch College of Business Sacred Heart University Fairfield, Connecticut USA Karsten Staehr Department of Economics and Finance Tallinn University of Technology Tallinn, Estonia ISSN 2198-7246 ISSN 2198-7254 (electronic) Springer Proceedings in Business and Economics ISBN 978-3-319-54884-5 ISBN 978-3-319-54885-2 (eBook) DOI 10.1007/978-3-319-54885-2 Library of Congress Control Number: 2017939566 © Springer International Publishing AG 2017 This work is subject to copyright All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed The use of general descriptive names, registered names, trademarks, service marks, etc in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations Printed on acid-free paper This Springer imprint is published by Springer Nature The registered company is Springer International Publishing AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland Preface This volume presents papers from the 2nd Wrocław International Conference in Finance held at Wrocław University of Economics on September 27–28, 2016 We have sought to assemble a set of studies addressing a broad spectrum of recent trends and issues in finance, particularly those concerning markets and institutions in Central and Eastern European countries In the final selection, we accepted 28 of the papers that were presented at the conference Each of the submissions has been reviewed by at least two anonymous referees, and the authors have subsequently revised their original manuscripts and incorporated the comments and suggestions of the referees The selection criteria focused on the contribution of the papers to the modern finance literature and the use of advanced analytical techniques The chapters have been organized along the major fields and themes in finance, i.e the econometrics of financial markets, stock market investments, macrofinance, banks and other financial institutions, public finance, corporate finance and household finance The part on the econometrics of financial markets contains seven papers The paper by Ewa Dziwok investigates some liquidity measures using data from the Polish market The paper by Agata Kliber analyses the impact of sovereign CDS on other instruments in financial markets The paper by Paweł Kliber examines the factors influencing overnight interest rates on the Polish interbank market Blanka Łe˛t studies in her paper whether the listings of natural gas prices in different derivative markets are linked Paweł Miłobe˛dzki examines whether the US dollar, the pound sterling, the Swiss franc and the Japanese yen are hedges or safe havens for Polish stocks and bonds Marta Chylin´ska and Paweł Miłobe˛dzki provide an application of a VEC DCC-MGARCH model for copper futures The paper by Piotr Płuciennik and Magdalena Szyszko presents an analysis of the dependences between inflation expectations extracted from inflation-linked swaps and the exchange rate, oil prices and the interbank rate The part on stock market investments contains four papers The paper by Agata Gluzicka applies the risk parity idea to the portfolios of stocks on the Warsaw Stock Exchange Sabina Nowak in her paper uses modified versions of models by Fama v vi Preface and French to include order imbalance factors The paper by Joanna Olbrys´ studies the interaction between market depth and market tightness on the Warsaw Stock Exchange In their paper Paulina Roszkowska and Łukasz Langer investigate mispricing in equity markets by studying abnormal excess returns determined by classical and modern asset pricing models The part on macrofinance contains five papers The paper by Małgorzata Iwanicz-Drozdowska and Paweł Smaga presents an analysis of factors influencing the development of financial systems in 40 countries The paper by Marta Karas´ and Witold Szczepaniak discusses an alternative method for calculating the CoVaR of the banking system In their paper Darko Lazarov, Tanja Lakovic and Emilija Miteva-Kacarski investigate the influence of the quality of financial information on the development of stock markets in 38 countries The paper by Magdalena Ligus and Piotr Peternek examines the preferences of home buyers in relation to urban environmental attributes Małgorzata Olszak and Iwona Kowalska study the effect of macroprudential policies and microprudential regulations on the sensitivity of leverage and liquidity-funding risks to the business cycle The part on banks and other financial institutions contains five papers The paper by Beata Lubinska presents a model of the optimization used for management of banking books Marta Małecka investigates VaR model testing for no-failure cases The paper by Helmut Pernsteiner and Jerzy We˛cławski contains an analysis of relationship banking in Poland Alicja Wolny-Dominiak analyses the prediction of total loss reserves in non-life insurance company by using a generalized linear model In their paper Ewa Wycinka and Tomasz Jurkiewicz investigate the use of a mixture cure model for a sample of consumer credit accounts of a Polish financial institution The part on public finances contains three papers Elena Querci and Patrizia Gazzola present an analysis of a model of health care providing low costs and high value The paper by Petra Ja´nosˇ´ıkova´ and Radka MacGregor Pelika´nova´ analyses the real estate transfer tax in different EU countries The paper by Tomasz Skica, Jacek Rodzinka and Rusłan Harasym contains an analysis of the impact of the financial policy of local government units on the development of entrepreneurship The part on corporate finance contains two papers Julia Koralun-Berez´nicka examines how the capital structure of companies in 13 EU countries depends on the firm size and debt maturity The paper by Elz˙bieta Rychłowska-Musiał describes investment decision rules using real options theory The part on household finance contains two papers Katarzyna Kochaniak analyses the risk profiles of household financial asset portfolios and their determinants in 15 euro area countries The paper by Beata Lewicka contains the analysis of factors which have a significant impact on having a consumer credit or a mortgage loan among people over the age of 50 We wish to thank the authors for making their studies available for our volume; their collegial, professional efforts and research inquiries made this volume possible We are also indebted to the anonymous referees for providing insightful reviews with many useful comments and suggestions Preface vii In spite of our intention to address a wide range of problems pertaining to financial markets, institutions and business organizations, we recognize that there are myriad issues that still need to be researched We hope that the studies included in our volume will encourage further research and analyses in the interesting field of modern finance Wrocław, Poland Fairfield, CO Tallinn, Estonia December 23, 2016 Krzysztof Jajuga Lucjan T Orlowski Karsten Staehr Contents Part I Econometrics of Financial Markets Chosen Measures for Pricing of Liquidity Ewa Dziwok Not as Black as Is Painted? Influence of sCDS Market on Domestic Financial Markets Before and After the Ban on Naked sCDS Trade Agata Kliber 11 Determinants of the Spread Between POLONIA Rate and the Reference Rate: Dynamic Model Averaging Approach Paweł Kliber 25 World Natural Gas Markets: Characteristics, Basic Properties and Linkages of Natural Gas Prices Blanka Łe˛t 35 Are Major Currencies Hedges or Safe Havens for Polish Stocks and Bonds? Paweł Miłobe˛dzki 45 Copper Price Discovery on COMEX, 2006–2015 Marta Chylin´ska and Paweł Miłobe˛dzki A Copula Approach to Backward-Looking Factors in Market Based Inflation Expectations Piotr Płuciennik and Magdalena Szyszko Part II 57 69 Stock Market Investments Risk Parity Portfolios for the Grouped Stocks Agata Gluzicka 81 ix x Contents Order Imbalance Indicators in Asset Pricing: Evidence from the Warsaw Stock Exchange Sabina Nowak 91 Interaction Between Market Depth and Market Tightness on the Warsaw Stock Exchange: A Preliminary Study 103 Joanna Olbrys´ Investment Opportunities in the WSE: Bull Versus Bear Markets 113 Paulina Roszkowska and Łukasz K Langer Part III Macrofinance Development of Financial Systems in 1995–2014: A Factor Analysis 125 Małgorzata Iwanicz-Drozdowska and Paweł Smaga Measuring Systemic Risk with CoVaR Using a Stock Market Data Based Approach 135 Marta Karas´ and Witold Szczepaniak The Quality of Financial Information and Stock Market Development: A Panel Data Study for the European Economies 145 Darko Lazarov, Tanja Lakovic, and Emilija Miteva Kacarski Impacts of Urban Environmental Attributes on Residential Housing Prices in Warsaw (Poland): Spatial Hedonic Analysis of City Districts 155 Magdalena Ligus and Piotr Peternek Macro- and Microprudential Regulations and Their Effects on Procyclicality of Solvency and Liquidity Risk 165 Małgorzata Olszak and Iwona Kowalska Part IV Banks and Other Financial Institutions Balance Sheet Shaping Through Decision Model and the Role of the Funds Transfer Pricing Process 183 Beata Lubinska Testing VaR Under Basel III with Application to No-Failure Setting 195 Marta Małecka Factors of Influence on Relationship Banking of Polish Firms 203 Helmut Pernsteiner and Jerzy We˛cławski Bootstrap Mean Squared Error of Prediction in Loss Reserving 213 Alicja Wolny-Dominiak 306 K Kochaniak Table The predicted structure of an average financial asset portfolio (in %) of a household characterised by NW from the class in individual countries and the Eurozone AT 90 BE 72 20 CY 51 38 11 DE 64 24 12 ES 80 11 FI 82 10 FR 80 14 GR 97 IT 85 12 LU 73 19 MT 84 12 NL 61 34 PT 91 SI 77 12 11 SK 85 10 EA 81 12 1st column: 1—S/TFA; 2—RS/TFA; 3—R/TFA Table The classes of NW with the greatest differences (in p.p.) in portfolio structure in comparison to class AT BE CY DE ES FI FR GR IT LU MT NL PT SI SK EA S/TFA (À21.85) (À37.61) (À13.95) (À23.26) (À46.76) (À32.09) (À53.42) (À12.25) (À34.88) (À22.02) (À35.44) (À21.38) (À16.05) (À29.92) (À8.67) (À34.85) RS/TFA (11.86) (9.66) (10.86); (À11.60) (8.11) (5.25) (9.31) (29.09) (5.42) (19.08) x (26.81) (11.01) (7.09) x x (13.97) R/TFA (9.99) (28.78) (17.33) (15.15) (42.86) (22.79) (24.33) (6.83) (15.80) (2.77) (15.03) (18.65) (8.96) (19.97) (6.48) (20.89) Numbers preceding parentheses show the classes of NW with the greatest changes in portfolio structure in comparison with NW from class The scale of change is reported in parentheses (in p p.) redirecting their interests primarily towards risky items The shares of relatively safe assets of households from classes 2–5 were also subject to increase However, regarding classes 3–5, they kept a shorter distance from that in a portfolio of a household from class 1.2 The above results led to the conclusion that a significant appetite for risky investments was demonstrated along with the multiplication of wealth in the Eurozone Analysing household investment preferences in individual countries, it can be found that the substitutability of safe assets by the remaining categories was present in all of them However, the results also revealed opposing tendencies towards relatively safe assets in the Cypriot sample, between a household with NW from class and a household with NW from class The first one was better equipped with relatively safe items than the basis for comparison (higher The changes in the predicted structure of a portfolio (in p.p.) caused by increasing NW from class to 5: S: À10, À17, À23, À35; RS—6, 8, 11, 14; R—4, 9, 12, 21 Does a Household’s Wealth Determine the Risk Profile of Its Financial 307 Table The predicted structure of an average financial asset portfolio (in %) of a household characterised by TRA, TFA, and T_LIAB from the class in individual countries and the Eurozone AT 92 BE 77 16 CY 61 28 11 DE 69 20 11 ES 83 8 FI 86 10 FR 82 13 GR 98 IT 90 LU 80 14 MT 90 NL 74 22 PT 95 SI 85 10 SK 90 5 EA 84 10 1st column: 1—S/TFA; 2—RS/TFA; 3—R/TFA share by 11 p.p.), while a household belonging to class possessed them of lower value (lower share by 12 p.p.) In Austria, Finland, France, Germany, Greece, Italy and Portugal, the greatest differences in portfolio structures occurred between those predicted for class and class (representing the wealthiest households) However, in some of them (Austria, France, and Italy) the decline of interest in safe assets has motivated a typical household to greater engagement in relatively safe assets rather than risky ones In the remaining countries (Germany, Finland, Greece, Portugal, and Spain) a household from the richest class was focussed the most on risky components of its portfolio Version of the model assumes the influence of the values of total financial assets (TFA), total real assets (TRA) and total liabilities (T_LIAB) on the structure of household portfolios The adoption of three independent variables allowed to obtain extended results which are partially presented in Table Detailed results are discussed in the text The basis for comparison remained a household, characterised by the values of all covariates from the lowest range (class 1) The proportions of its safe, relatively safe, and risky assets varied between the member states (Table 5) In the Eurozone, possessing real or financial assets of the highest values tended to downplay the importance of safe assets and raised the importance of the other categories (Table 6) Thus, a household holding them was more oriented to relatively safe and risky assets than the basis for comparison However, precise outcomes showed that increasing TRA (class by class) made risky assets more attractive, while increasing TFA—those relatively safe The greater risk exposure was noted for a household with liabilities exceeding the debt level established for class It means, the more actively debt was used as a funding source, the more risk-oriented investments in financial assets became However, detailed results disclosed that increasing T_LIAB caused a growth rather in the shares of relatively safe assets than risky ones.3 In individual countries, portfolio structure has not maintained itself in such an unambiguous manner In Finland, France, and Italy any value of TRA exceeding class resulted in a reduction of the share of deposits and favoured the other asset categories, but the greatest changes occurred between portfolios of households with real assets from class and class From the impact of the value of TFA on the portfolio’s structure, a general conclusion could be drawn: the value beyond class stimulated greater interest in relatively safe or risky assets Complete results for Austria, Belgium, Finland, France, Germany, Italy, Luxembourg, Malta and Portugal prove that this interest was focussed on both of Average increase (in p.p.) of RS by: 5, 6, 6, 5, while of R by: 1, 1, 1, 308 K Kochaniak Table The classes of TFA, TRA, T_LIAB with the greatest differences in portfolio structure in comparison to class AT BE CY DE ES FI FR GR IT LU MT NL PT SI SK EA S/TFA TFA (À) (À) (À) (À) (À) (À) (À) (À) (À) (À) (À) (À) (À) (À) (À) (À) TRA (+) (À) x (+) (À) (À) (À) (À) (À) x (À) (+) x x (+) (À) T_LIAB (À) (À) (À) (À) (À) (À) (À) (À) (À) (À) (À) (À) (À) (À) (À) (À) RS/TFA TFA TRA (+) (À) (+) (+) x (À) (+) (À) (+) (+) (+) (+) (+) (+) (+) x (+) (+) (+) (À) (+) (+) (+) (À) (+) (+) (+) (À) (+) (À) (+) (+) T_LIAB (+) (+) (+) (+) (+) (+) x x (+) (+) (+) (+) (+) (+) (+) (+) R/TFA TFA (+) (+) (+) (+) (+) (+) (+) (+) (+) (+) (+) (+) (+) x (+) (+) TRA (À) (+) (+) x (+) (+) (+) (+) (+) x (+) (+) (À) x x (+) T_LIAB (À) x x (À) (+) (À) (+) (+) (+) x x (À) (+) (+) x (+) Numbers preceding parentheses show the classes of TRA, TFA, and T_LIAB with the greatest changes in portfolio structure in comparison to class 1; (À) decreased share of portfolio’s part, (+) increased share of portfolio’s part their types In the case of T_LIAB, detailed outcomes confirmed positive influence of all debt classes above class on the shares of relatively safe assets in portfolios of households residing in Belgium, Cyprus, Germany, Luxembourg, Portugal, and Spain Moreover, the detailed results from version of the model revealed sub-groups of countries with the following consequences of the values of TFA, TRA and T_LIAB from classes 2–5 for the portfolio structure: (1) Belgium, France, Greece, Italy, Malta, and Spain: generally, lower proportions of safe assets were accompanied by increased proportions of relatively safe and risky assets The Eurozone portfolio’s profile corresponded to this structure; (2) Austria, Germany, the Netherlands, and Slovakia: total financial assets (TFA) from classes 2–5 led to a lower proportion of safe assets in the portfolio and higher shares in two other asset categories In contrary, total real assets (TRA) from classes 2–5 strengthen the position of safe assets The greatest sensitivity of the structures of financial asset portfolios appeared at different values of total financial assets (TFA), total real assets (TRA) and total liabilities (T_LIAB) in the Eurozone member states In Portugal, notable changes could be observed when the values of covariates were assigned to classes 2–3 On the other hand, there was a group of countries, where the most prominent differences occurred when the values of covariates were from classes 4–5 This lesser Does a Household’s Wealth Determine the Risk Profile of Its Financial 309 sensitivity of portfolios’ risk profiles could be observed for the populations of Cyprus and Italy, and with small exceptions of Finland, France, Italy, Luxembourg, and Malta The same could be concluded for the Eurozone as a whole Conclusions Despite the development of the financial market since the 1990s, deposits appeared as a standard component of households’ portfolios Moreover, in almost all the countries analysed they emerged as dominant Households’ investment preferences, expressed by average portfolio structure, could not be perceived as uniform in the entire group of countries However, it was possible to identify three sub-sets of member states with similarities in this respect Analysing the impact of net wealth on the structure of households’ portfolios in the Eurozone, the results from version of the fractional multinomial logit model showed that a higher level of net wealth led to increases in the proportion of both risky and relatively safe assets However, risky ones were present in the focus of interest of more affluent households The results from version of the model considering the whole group of countries suggested a positive impact of the greater value of total real assets on the shares of risky assets in portfolios, while total financial assets on the proportions of relatively safe ones However, in certain countries increasing total real assets led to the growth in the shares of safe assets According to the results derived from both variants of the model for separate countries, domestic preferences used to determine the structure of households’ portfolios There was no compliance in the importance of the safe, relatively safe and risky assets as well as in the direction of the impact of separate covariates on portfolio structure Acknowledgements The paper presents the results of a study carried out under a research project financially supported by the National Science Centre in Poland (grant number: DEC-2013/11/D/ HS4/04056) The analysis is based on data provided by the Eurosystem Household Finance and Consumption Survey References Bilias Y, Georgarakos D, Haliassos M (2008) Equity culture and the distribution of wealth Netspar DP 2008–010, pp 1–43 Buis ML (2008) Fmlogit: stata module fitting a fractional multinomial logit model by quasi maximum likelihood Statistical software components Boston College Department of Economics, Boston Christelis D, Georgarakos D, Haliassos M (2011) Differences in portfolios across countries: economic environment versus household characteristics http://ssrn.com/abstract¼1089802 Accessed Feb 2016 Cocco JF (2005) Portfolio choice in the presence of housing Rev Financ Stud 18(2):535–567 310 K Kochaniak Du Caju P (2013) Structure and distribution of household wealth: an analysis based on the HFCS NBB Econ Rev 9:41–62 ECB (2013) The eurosystem household finance and consumption survey—results from the first wave ECB Stat Pap Ser 2:1–112 Guiso L, Paiella M (2008) Risk aversion, wealth, and background risk J Eur Econ Assoc (6):1109–1150 Guiso L, Haliassos M, Jappelli T (2002) Household portfolios: an international comparison In: Guiso L, Haliassos M, Jappelli T (eds) Household portfolios MIT Press, Cambridge, pp 1–24 Haliassos M (2002) Stockholding: recent lessons from theory and computations In: Guiso L, Haliassos M, Jappelli T (eds) Stockholding in Europe Palgrave Macmillan Publishers, London, pp 30–51 Kukk M (forthcoming) How does household debt affect financial asset holdings? Evidence from euro area countries Stud Econ Financ Mullahy J (2011) Multivariate fractional regression estimation of econometric share models UCD Geary Institute Discussion Paper Series WP2011/33, pp 1–45 Murteira JM, Ramalho JJ (2013) Regression analysis of multivariate fractional data CEFAGE-UE Working Paper, 2013/05, pp 1–43 Teppa F, Porpiglia A, Ziegelmeyer M, Le Blanc J, Zhu J (2015) Household saving behaviour and credit constraints in the euro area ECB WP 1790, pp 1–62 Supporting Family to Their Utmost— People’s over the Age of 50 Attitudes to Borrowing Beata Lewicka Abstract The aim of this paper is to indicate factors which have a significant impact on having a consumer credit or a mortgage loan among people over the age of 50 The article contains results of a research which was carried out in the period from April to May 2016 Respondents of this research were people over the age of 50 who live in lubelskie region (Poland) The sample reflects the proportion of age groups in tested population (stratified sampling) In order to reach the goal of this paper and verify stated hypotheses, statistical tests of significance and binary logistic regression were used The conducted study revealed that supporting family is a strong motive, which has an impact on having or not consumer or mortgage loan Moreover, the willingness to help family financially is much more important than any other financial reason or obstacle At the same time, the level of trust in family is not important when people borrow money for them Introduction Elderly people in Poland constitute a group which is said to be particularly vulnerable to financial exclusion This situation is partly due to the fact that they often face difficulties in staying up-to-date with financial products (Smyczek and Matysiewicz 2014) Furthermore, some of them not present sufficient level of financial literacy (Kuchciak 2014) Nevertheless, other researches revealed that it is not only the demand-side problem Financial exclusion and consequently phenomena of unbanking or underbanking of people over the age of 50 may be caused by poor and not relevant offers available on the financial market (Ziemba et al 2014) Moreover, older people frequently face limited access to loan instruments (Buk and Pustiowska 2013), especially mortgage loan, which cost should be calculated taking into account the possibility of creating income on retirement (Polish Financial Supervision Authority 2013) B Lewicka (*) Maria Curie-Sklodowska University in Lublin, Lublin, Poland e-mail: beata.lewicka@umcs.pl © Springer International Publishing AG 2017 K Jajuga et al (eds.), Contemporary Trends and Challenges in Finance, Springer Proceedings in Business and Economics, DOI 10.1007/978-3-319-54885-2_28 311 312 B Lewicka However, despite of all mentioned barriers, borrowing activity of “baby boomers” generation is similar to younger generation “X” According to Bureau of Credit Information (BIK) statistics, in 2015, 54.4% of people born between the years 1948 and 1967 had an active loan In comparison, for people born between the years 1968 and 1981 this borrowing activity rate was 58% What is the most surprising, among the eldest citizens (born before 1948), 38.2% were still active borrowers The reasons why Polish people over the age of 50 borrow money are not sufficiently examined in academic literature Nevertheless, some problems are common for all European baby boomers, no matter from which region they come from The reasons of borrowing can be diverse: having prior commitment, experiencing unexpected events, protecting existing assets or facing problems with making ends meet (Finney 2013) Surprisingly, despite the fact, that so many of Polish elderly people are indebted, they are still generous for their families The latest data collected by the Central Statistical Office (2016) has revealed that Polish pensioners in 2015 spent more than billion zlotys on purposes connected with supporting their relatives (an average pensioners’ household spent on this purpose yearly 670 zlotys per capita) Intergenerational transfers can be divided into two groups: inter vivos –between living generations and bequests—transfers of wealth made after death (Albertini and Radl 2012) Another difference is that bequests are usually equally distributed across all children in the family, while inter vivos transfers are made unevenly and are more directed to the more needy children (Kohli and Künemund 2003) The latter also differ across countries and regions For an instance, in Southern Europe, where adult children prolong their co-residence with parents, economic support from them is smaller than in Scandinavian countries, where independent living in early adulthood is preferred Continental European countries are classified as middle-of-the-road (Albertini and Kohli 2013) Poland case is similar to Continental regime Results from Polish sample in SHARE project revealed that children are mostly supported financially by their parents, whereas nonfinancial transfers flow from younger to older generation (Nicin´ska and Kalbarczyk 2009) Using borrowing instruments, despite alleged financial exclusion as well as reported tendencies of increasing spending on support for kinship led to assumption that in Poland intergenerational transfers are partially “leveraged” The aim of this article is to find out factors which have a significant impact on having consumer or mortgage loan In this article three hypotheses are stated: H1: Among people over the age of 50, supporting family is one of the motives underlying a decision to take out consumer credit H2: Among people over the age of 50, supporting family is one of the motives underlying a decision to take out mortgage loan Supporting Family to Their Utmost—People’s over the Age of 50 313 H3: The role of trust in one’s family is not significantly important when making decision about borrowing money from banks to support one’s relatives Nowadays, issues touched in this article seem to be especially important taking into consideration population ageing process and all challenges which derive from demographic changes in Poland and, more broadly, in all developed countries Despite the fact that debt problems of households and their economic and social determinants are well-covered in subject literature (i.e Cynamon and Fazzari 2008; Debelle 2004; Georgarakos et al 2014), there is no specific research on borrowing of people over the age of 50 and role of family in their approach towards it Hence, this article contribution is to begin a discussion about this difficult, partially economic and partially social issue Methods and Data The data used in this paper comes from the research which was carried out in the period from April to May 2016 The respondents of this research were people over the age of 50 who live in lubelskie region The minimum sample size (384) was calculated for margin error equal 5% and 95% confidence level The sample (386 answers) reflects the proportion of age groups in tested population (stratified sampling) Nevertheless, this sample size is still a limit of the work, as larger sample would give more reliable results with greater power and precision The chi square test revealed that number of observations in each age category was not significantly different from the actual structure of the population for tested region Respondents were asked to complete a paper-based questionnaire, which included questions about their financial literacy, borrowing habits and experiences from financial markets In order to confirm stated hypotheses (H1, H2) the binary logistic method was used Reasons why this method was applied are that many variables in suggested model were dichotomous (Tranmer and Elliot 2008) and its interpretation is clear and readable Variables which were taken into consideration were divided into four groups: the use of a financial product/service, types of experienced difficulties with access to financial services, main reasons for borrowing money and respondent characteristics (Table 1) Selection of the variables included into model was carried out in five stages (Bursac et al 2008, Canchola et al.): (1) descriptive analyses of variables, (2) univariable analysis of each variable, (3) testing collinearity between candidate variables (testing Spearman correlation coefficients), (4) multivariable analysis (using backward elimination method), (5) evaluating all estimated models with Hosmer-Lemenshow Goodness-Of-Fit 314 B Lewicka Table Description of used variables Variable-name Name of variable abbreviation Coding method Declaration of using Checking bank account S_CBA 0-no, 1-yes Savings bank account S_SBA Bank deposit S_BD Brokerage account S_BA Investment fund account S_IFA Individual retirement account S_IRA Credit card S_CC Consumer loan S_CL Mortgage loan S_ML Main reasons for borrowing money Unexpected expenses M_UE 0-no, 1-yes Supporting family M_SF Temporary financial problems M_TF Lack of money to meet basic needs M_BN Spending more than earn M_SM High cost of expense M_CE Types of experienced difficulties with access to financial services Problems with meeting all necesD_NR 0-never, 1-very rarely, 2-rarely, sary requirements 3-sometimes, 4-often, 5-very often Ambiguous or unclear offer D_AO Too much choice D_CH Too small font in presented D_SF documents Too long queues in agencies D_LQ Incompetent staff D_IS Lack of amenities for older people D_LA in agencies Lack of suitable offer D_LO Hidden costs D_HC Too complex language in D_CL presented documents Respondent characteristics Sex SEX 0-Female, 1-Male Age AGE 1-50–55 years old; 2-56–65, 3-66–75, 4-over 75 years old Declared financial literacy level FLL 1- much lower, 2-lower, 3-similar, 4-higher, 5-much higher Self-evaluation of financial situaEFS 1-very bad, 2-bad, 3-not bad, not good, tion (regarding to average) 4-good, 5-very good Monthly earnings per capita ME Standarised for model Supporting Family to Their Utmost—People’s over the Age of 50 315 Table Number of replies estimated for categorical variables Binary variable S_CBA Yes No S_SBA Yes No S_BD Yes No S_BA Yes No S_IFA Yes No S_IRA Yes No S_CC Yes No S_CL Yes No S_ML Yes No M_UE Yes No M_SF Yes No M_TF Yes No M_BN Yes No M_SM Yes No M_CE Yes No SEX Male Female Having taken out consumer credit No Yes 245 27 110 88 267 29 141 214 24 10 345 31 28 327 29 20 335 31 56 299 23 Χ Χ Χ Χ 20 335 26 84 17 269 14 34 13 319 18 48 305 23 10 343 29 349 30 79 274 23 134 17 218 17 Having taken out mortgage credit No Yes 252 20 109 84 277 19 144 217 21 352 24 28 333 23 20 341 25 57 304 18 26 335 20 Χ Χ Χ Χ 92 267 16 40 319 18 54 305 23 12 347 25 355 24 71 16 288 138 13 220 12 x—factors not used as explanatory variable in selected model In order to select variables which can be used in estimated models, descriptive analyses of collected data were carried out and presented in Tables and Categorical variables were examined if there are any zero frequencies, whereas for ordinal and linear variables their variability levels were tested Thereafter, univariable analyses helped to indicate associations of single variables with the outcome, which resulted in exclusion of unimportant variables— those having insignificant Wald test and p-value cut-off point of 0.25 (Bursac et al 2008) Next, backward elimination method allowed to isolate those variables, 316 B Lewicka Table Means and standard deviations calculated for ordinal and linear variables Variable D_NR D_AO D_CH D_SF D_LQ D_IS D_LA D_LO D_HC D_CL AGE FLL EFS ME Having taken out consumer credit No Yes Mean SD Mean SD 1.320 1.681 1.097 1.640 2.386 1.707 1.903 1.700 2.252 1.766 2.065 1.413 3.009 1.838 2.806 1.621 2.220 1.671 1.903 1.720 1.620 1.557 1.258 1.527 1.484 1.632 1.032 1.472 1.920 1.615 1.581 1.311 2.783 1.886 2.516 1.411 3.239 1.715 3.161 1.675 2.310 1.000 2.129 0.991 2.847 1.054 3.387 1.022 2.997 0.841 3.355 0.915 1161.12 730.83 1092.08 703.72 Having taken out mortgage credit No Yes Mean SD Mean SD 1.329 1.691 920 1.441 2.346 1.728 2.360 1.440 2.248 1.769 2.080 1.256 3.031 1.803 2.440 2.022 2.225 1.687 1.760 1.451 1.624 1.572 1.120 1.236 1.496 1.647 760 1.012 1.910 1.620 1.640 1.150 2.730 1.871 3.200 1.528 3.249 1.726 3.000 1.472 2.346 0.997 1.560 712 2.877 1.072 3.080 862 3.022 0.852 3.080 862 1148.14 730.59 1260.33 695.74 which explain why people over the age of 50 borrow money from financial institutions Finally, in an attempt to find out if the level of trust in family is not significantly different for people who support their families with their loans and those who not that (H3), the Mann-Whitney U test was carried out This non-parametric test was chosen firstly because, the trust in family was measured on Likert-like-type rating scale, which should not be treated as interval scale (Jamieson 2004) and secondly, because the independent variable was not normally distributed Results The results of the conducted research revealed, that among many aspects which describe borrowers from lubelskie region over the age of 50, those connected with motivation of borrowing are the most significant The first model, which presents factors influencing on taking out consumer credit, indicates that supporting family (M_SF) and occurrence of unexpected expense (M_EU) hold a very strong meaning Those who claim to support their families are five times (5.172) more likely to have a consumer credit (ceteris paribus) Whereas those who mention an unexpected expense as a reason of borrowing are only two times (2.211) more likely to get a consumer loan from bank or other institution Another significant factor estimated in the model is holding a saving bank account (S_SBA) The probability of having a consumer credit for those who claim to have at least one saving account is 82% lower than for those who not have it The last significant factor is a Supporting Family to Their Utmost—People’s over the Age of 50 317 declared level of financial literacy (FLL) The likelihood of taking out consumer credit grows with the self-evaluated financial literacy level The first model was tested with Hosmer-Lemenshow Goodness-Of-Fit (p-value 0.556) The value of Nagelkerke R2 estimated for this model means that it is able to explain 24.4% of the variation based on the given variables (Table 4) The second model indicates which factors impact on having mortgage loan by people who are over the age of 50 (Table 5) Again, those who support their family are more likely to have a mortgage credit (2.28 times more, ceteris paribus) Nevertheless, the likelihood increases the most for people who claim that too high cost of expense is their main borrowing motive (likelihood is 7.5 times higher) Contrary to the consumer credit model, for this type of loan, three barriers in access seem to be significant Firstly, lack of suitable offer (M_LO) and secondly, small font in all necessary documents (M_SF) If a respondent experiences those obstacles more frequently, the likelihood of holding a mortgage is smaller Third significant barrier works differently—for people who experience hidden costs more often, the likelihood of having a loan increases The goodness of fit for this model was assessed with Hosmer-Lemenshow test (p¼0.485) The value of Nagelkerke R2 estimated for this model means that it is able to explain 24.4% of the variation based on the given variables Another tested hypothesis (H3) assumes that the level of trust in family is not important for people for whom family support is one of the main motivation for borrowing Differences between level of trust in family for those who borrow Table Logit model for borrowers over the age of 50—consumer credit Variable S_SBA M_UE M_SF FLL Constant Beta coefficient À1.717 1.167 1.820 520 À4.658 Standard error 777 410 439 203 749 Wald 4.880* 8.101** 17.216** 6.532* 38.709** Exp(B) 180 3.211 6.172 1.682 009 *p-value