Differences in the Level and Structure of Household Indebtedness in the EU Countries

The article presents the outcomes of dynamic, cross-section analysis of the differences in the level and structure of household indebtedness in EU countries and the range of problems with debt service between 2005 and 2009. Statistical data from of the EU Commission, obtained in the periodic research of people’s incomes and living conditions (EU-SILC) and the ECRI - European Credit Research Institute, was used in the research, as well as the method of multi-dimensional analysis (cluster analysis and k-means method), enabling to classify EU households according to the features of their indebtedness. Moreover, in order to define the quantitative relationships between the level and structure of households’ indebtedness and the frequency of repayment problems, tools such as correlation analysis and stepwise regression, enabling the description of its strength and direction of influence of selected variables on repayment problems.


Introduction
It is possible to notice that there has been a substantial rise in households' indebtedness in the EU over the last several years, measured as different categories, both in absolute terms and as a share in incomes and GDP (Lilico, 2010). A rise of household indebtedness measured as a ratio of debt to disposable income is especially common in advanced economies (Bloxham & Kent, 2009). Credits enable households to sustain the consumption growth, simultaneously leading to the decline in household saving rate (Barba & Pivetti, 2009).
The very phenomenon of households becoming indebted is natural and common in developed coun-tries with modern financial systems. Over the last decades the attitude to a credit has changed and it has become a part of a modern consumer society (Lea, Webley & Walker, 1995). Households accustomed to live on credit, and they treat it as an ordinary resource in a household economy (Raijas, Lehtinen & Leskinen 2010). According to Dynan (2009) the increase in household borrowing had widespread gradually over the time across different demographic groups.
The median of debt-to-income ratio had risen for all demographic groups (taking into account factors such as: age, education, income, housing tenure). A distinction should therefore be made between two notions: indebtedness and over-indebtedness (Russell, Maître & Donnelly, 2011). Joyce (2003) defined indebtedness as "a commitment to repay moneys which a debtor has borrowed and used" (as cited in Law Reform Commission, 2009,

Differences in the Level and Structure of Household Indebtedness in the EU Countries
Differences in the Level and Structure of Household Indebtedness in the EU Countries p. 28). The definition views getting indebted as a positive and inevitable consequence of crediting, beneficial for both the whole society and particular individuals.
The majority of credits are repaid without problems and brings benefits to all parties of the agreement.
Credit availability is one of the major factors of economic development in the contemporary credit society, but also a factor of household financial sustainability (Niemi-Kiesiläinen & Henrikson, 2005). The theory of economics views debt as an instrument stabilizing households' consumption over time (Gumy, 2007).
However, a dynamic increase of both the volume and value of credits granted to households can contribute to the creation of over-indebtedness if households' finances are mismanaged and if their financial awareness is low and financial education is inappropriate (Bywalec 2009;Świecka, 2008;). The key question for the policy makers should concern the ability of households to service their current level of debt (Brown & Taylor, 2008).
Over-indebtedness is a relatively new notion, not having a single correct interpretation. A number of attempts have been made to define it in Europe as well as describe the phenomenon of over-indebtedness. Since Two approaches are represented by researchers dealing with households' indebtedness; one of these dealing with the description what should be understood as over-indebtedness and the other one, describing the indicators used in order to assess it. In reality, there is no single statistics to assess the multi-dimensional phenomenon of households' over-indebtedness (European Commission, 2008). At the same time, three basic models, compiled by the European Commission, aimed to measure over-indebtedness can be distinguished: the objective model, the subjective model and the administrative model (Betti et al., 2007).
Objective ratios are the measurable, based on quantitative data. They include such measures as the consumption/income ratio, the debt/asset ratio or the debt payment/income ratio, describing the possibilities of debt repayment. (Betti et al., 2007).
Objective measures also include the ratio based on arrears (arrears indicator). A household is considered to be over-indebted if it has arrears in credit repayment and / or paying liabilities connected with flat maintenance exceeding three months (Fondeville, Ozdemir & Ward, 2010).
The subjective model assumes that household members know their own financial situation the best.
Thus, subjective measures take into account the views of families concerning debt repayment problems.
A household is over-indebted if it assumes that debt repayment constitutes too large a financial burden (Kempson, 2002;Gumy, 2007).
Administrative measures of over-indebtedness, in turn, are based on official data concerning the formal procedures of acting in over-indebtedness cases.
The aim of the present paper is to classify the EU countries according to the level and structure of households' indebtedness as well as according to the problems with repaying obligations. Also, an attempt is made to answer the question how the structure of indebtedness influences debt repayment problems.

Source materials and methodological assumptions
The statistical data used in the present paper was obtained from the EC Report (Fondeville, Ozdemir & Ward, 2010). The report includes a number of statistics concerning the issues of households' indebtedness, basing on the data of the European Credit Research Institute (ECRI) 1 , Eurostat data on people's incomes and living conditions (EU-SILC) 2 and national accounts.
The research and analysis of the level and structure of indebtedness covered all countries except for Cyprus, Luxembourg and Malta, due to lack of data.
In order to classify the EU countries according to the level and structureof households' indebtedness as well as repayment problems, the methods of clustering high-dimensional data was used in the study. Thus, it was possible to describe a number of features simultaneously. Using the above-mentioned methods also made it possible to divide the collection of observations into relatively homogenous sub-collections, basing on the features describing the sub-collections (Poczta-Wajda, 2010). and structure and problems connected with debt service among households. The clusters are generally formed in order to minimize inside-group variance and maximize the variance between groups. In other words, the aim of the method is to form homogeneous groups, minimizing the variance in clusters and maximizing the variance among them (Stanisz, 2007). Moreover, the tools of analyzing correlation and regression were used in the study in order to identify the strength and directions of the relationships among the features describing the studied phenomenon.

The level of households' indebtedness in EU countries
One of the basic ratios used in the comparative analysis of households' indebtedness level among countries is the quotient of households' credit indebtedness and their disposable incomes (Lilico, 2010  Lithuania. The annual average proportion of households' debt in their disposable incomes was subject to a steady growth by approximately 25%. However, these countries were also characterized by relatively low proportions of indebtedness to incomes. In spite of the dynamics, indebtedness constituted only 33,3% of disposable incomes in Romania, which was the lowest value in the whole European Union. Indebtedness rates in Slovenia and Slovakia reached similar levels. These countries were, however, characterized by a much slower rate of credit debt increase compared to incomes. Credit indebtedness in Poland and Lithuania did not, in turn, exceed a half of households' disposable incomes during the whole studied period. However, as it has been stressed before, the dynamics of the changes in those two countries was really high. The indebtedness of Polish and Lithuanian households had more than doubled in proportion to incomes, which was five times faster a pace than in the EU on average.
To sum up, the quoted data unequivocally shows that there are considerable discrepancies among EU countries as regards the level of households' indebtedness and the dynamics of its changes. The households of the 'old' EU countries are generally more indebted, but their indebtedness increases more slowly. The households in Central and Eastern Europe or countries which accessed the EU after 2004 are, in turn, less indebted, but the dynamics of the changes resulting into a greater degree of indebtedness is much higher. It mainly results from the differences caused by tradition and the level of development of financial markets. Other causes include income conditionings, financial awareness in a given society and the hitherto model of households' functioning in totally different social and economic systems.

The structure of households' indebtedness in EU countries
Households' indebtedness should also be analyzed from the perspective of its typological structure, i.e. loans and mortgages indebtedness, indebtedness due to other obligations.
Differences in the Level and Structure of Household Indebtedness in the EU Countries  Credits and loans taken out for accommodation reasons, including the purchase, maintenance and repair of people's places of residence, constitute the greatest part of households' indebtedness in EU countries.
The proportion of the volume of mortgages to households' disposable incomes and also their share in the general structure of indebtedness as well as the dynamics of changes of both ratios are presented in Ta I  II  I  II  I  II  I  II  I  II  I     The last analyzed type of households' indebtedness is the indebtedness which does not belong to any of the categories of mortgages or consumption credits.
Its level and dynamics are presented in Table 4

Households' problems with regular repayment of obligations
The data presented above quite distinctively draws the scale of the problem of households' indebtedness, measured as the proportion of debt to disposable incomes.
It is difficult, however, to describe the influence of increasing indebtedness on the economic and financial situation in households, which is, to a large degree, influenced by payments connected with debt service.
For it is the impossibility to repay on time and not the level of indebtedness that mainly contributes to the deterioration of financial condition which can result in insolvency.
The range of the problem is reflected by the data shown in Table 5. Its analysis implies that the Bulgarians have the greatest problems with debt repayment as every third household declared difficulties in repaying at least one of the following obligations: mortgages, rent credits, other credits and municipal charges.
At the same time, as many as 25% of households in Greece and Romania acknowledged that they had problems repaying at least one liability.
The most systematic repayments were observed in the Netherlands, Czech Republic and Denmark where only 4% of households confessed to repayment problems.
The typology of households in EU countries according to the level and structure of indebtedness and the range of repayment problems Basing on the data from 2008 included in tables 1-5, particular EU countries were divided into 7 groups characterized by different level and structure of households' indebtedness as well as repayment problems.
The outcomes of the classification obtained by means of data clustering 3 are presented in Table 6.
The first group includes countries such as Ireland, Spain, Portugal, Sweden and Great Britain. It is a group of countries whose households are heavily indebted and in which credit indebtedness exceeded households' incomes. As regards indebtedness structure, mortgages constituted the dominating part, i.e. about 75% of the value of all credits. The households in question had, however, relatively fewer repayment problems, and only 7% declared them.
Greece is classified in group two as its only element.
It was characterized by much lower a rate of households' credit indebtedness in proportion to disposable incomes.
Credit indebtedness in Greece amounted to nearly 70% of households' disposable incomes. However, considerable repayment problems were observed in this country, as every fourth household declared serious problems repaying at least one obligation. It might be caused by consumption credit indebtedness, which amounted to almost ⅓ of the total value of all credits.
The next typological group consists of Belgium, Germany, Estonia, Latvia, Austria and Finland. It is a group of countries in which households are characterized by a medium level of indebtedness whose proportion to disposable incomes was lower than the average for the EU.
The value of the indebtedness ratio in this group amounted to 80%. It is worth stressing that the declared problems repaying liabilities concerned a lower number of households than the average number in the EU.
However, as regards households' indebtedness structure in these countries, the debts were constituted mainly by mortgages (70%), commonly believed to be served best.
Denmark and the Netherlands constitute a separate group. It is the most heavily indebted group in which the credit where households' indebtedness exceeds their disposable incomes more than twice. Such high level of indebtedness was mainly the consequence of mortgages and flat loans, which amounted to almost Differences in the Level and Structure of Household Indebtedness in the EU Countries Table 5. Percentage of households declaring arrears in repaying at least one liability in EU countries (data from 2008) Table 6. Classification of EU countries according to the level and structure of indebtedness and repayment problems -average percentages for particular indebtedness groups (data from 2008)

Country % of households with arrears (at least one liability)
Country % of households with arrears (at least one liability)

Quantity analysis of the influence of selected variables on the frequency of repayment problems in EU countries
In order to describe the strength and direction of the selected variables on the frequency of repayment problems, analysis of correlation and the method of stepwise regressions were used. The problem was analyzed in two versions and only based on data from 2008 for which it was possible because of the availability of data.
The following variables appear in the first option:  Table 7 presents the matrix of correlation between the analyzed variables. The analysis of correlation ratios shows a high and, in a vast majority of cases, significant relationship between the ratios of indebtedness structure and the frequency of problems with its service, measured as the percentage of households reporting arrears in repaying at least one liability.
In light of the above data, debt service problems were to the largest degree connected with financing households through consumption credits (X 3 , X 6 ). In this case they amounted to R=0.611 and R=0.824 respectively, which confirms the earlier observations that the more the share of these credit is in proportion to disposable incomes or total indebtedness, the greater the percentage of arrears. It also needs to be stressed that the degree to which the variability of the number of households reporting debt service problems is considerably higher in the model taking into account consumption credits (R2=68.03%).
It means that this particular kind of credits is, for EU households, the main source of financial problems.

Summary
Households' over-indebtedness is presently becoming a very important social and economic problem for both highly developed countries and emerging markets. However, the problem has a dual nature. As regards developed countries, in spite of a relatively good rate of repayment, the increasing level of households' indebtedness is gradually becoming a problem. On the other hand, in the remaining countries, the high dynamics of the increase, in spite of the still low level of households' indebtedness -is starting to overcome the pace of income rise. Although there are a number of reasons for it, in times of crisis, such phenomena as excessive consumption, lack of knowledge or limited financial awareness constitute the factors which particularly strongly affect households' financial condition, resulting, in a number of cases, in arrears and even insolvency. The conducted research also seems to emphasize the particular role of financial institutions in preventing those negative phenomena from happening. Easy access to credits and, particularly, consumption credits, has already led to serious problems in households, but also serious problems of the crediting financial institutions, which, as a consequence, resulted in a number of negative tendencies in the macro scale. One may imply that the world economic and financial crisis will enforce significant changes in this respect. Households are an integral element of each social and economic system, which means that the debt service problems that they encounter will negatively influence the whole economy in the long run. 3 The k-means method and statistical program, STATISTICA 9.0, were used.