IMPACT OF BANKING GROUP’S MANAGEMENT POLICIES AND COUNTRY ENVIRONMENT ON BANK’S MODEL OF OPERATION

The objective of this paper is to seek to identify which of two factors: (1) the banking group management policies, or (2) the country in which the bank operates, has higher impact on the bank’s model of operation measured in terms of three ratios: net fee and commission income compared to net interest income, impairment losses on loans compared to net interest income and net fee and commission income compared to personnel expenses. The analysis included 29 banks affiliated in four banking groups and nine CEE countries during the period 2014 and 2015. The results show that the country environment has much higher impact on banks’ operations than their management policies.


INTRODUCTION
Revenue from interests, fees and commissions is the most important source of the bank's income.Interest income, that i, money earned by lending out customers' deposits in various forms of loans, makes the most signi cant part of banks' income.Fee income is the revenue taken in by banks from account-related charges to customers.Fees and commissions consist mainly of fees charged on payment services, credit cards, services and fund management on behalf of legal entities and citizens, together with commissions from guarantees.Fee income and acquisition costs related to loan origination are usually capitalized and included in the e ective interest rate and recognized as interest income over the expected life of the loan.
is paper analyses the magnitude of net interest income and net fee and commission income of the banks that are members of one of the four banking groups (Banca Intesa, UniCredit, Rai eisen and Societe Generale) within nine Central and Eastern European (CEE) countries, the trend of the extent of net interest income (i.e. interest income less interest expenses) used to cover bad loan losses in 2014 and 2015, as well as the trend of net fee and commission income (i.e.fee and commission income less fee and commission expenses) used to nance personnel expenses.In other words, the paper analyses the e ciency by means of which certain segments of income nance certain costs.
e usual approach to banks' e ciency analysis is to compare various trends in "costs to income ratio" as an overall indicator of cost e ectiveness.Costs to income ratio measures income generated per monetary unit cost.e lower the value of this ratio, the better the performance of the bank.According to Rai eisen RESEARCH (Rai eisen, 2016), which analyses CEE banking sector in 2014 and 2015, the results of e orts of surveyed banks to slow down costs had di erent results over the period and by di erent banks.UniCredit is the only bank in CEE that was able to reduce the is study is focused on examination of the components of banks' income and expenses, starting from the idea that interest income represents "time value of money" used, while fee and commission income is "earned" by various non-lending activities.e ability of banks to nance personnel expenses, as part of total operating expenses, by net fee and commission income, as part of total income, may be an indicator of the extent of supreme knowledge, or supreme technology, or supreme management, or similar capabilities deployed as intellectual and human capital.According to the International <IR> framework, intellectual capital represents "organizational, knowledge-based intangibles, including "organizational capital" such as tacit knowledge, systems, procedures and protocols", while human capital represents "people's competencies, capabilities and experience, and their motivations to innovate, including … their loyalties and motivations for improving processes, goods and services …" (IIRC, 2013, p. 12).According to Ho mann, the e cient-structure hypothesis says that banks with superior management or production technologies have lower costs (Ho mann, 2011, p. 258).
Moreover, almost all banks make special e ort to further improve their capabilities by introduction of new technologies, especially in the area of digital banking due to minimum investment in xed assets and further opportunities for reduction of labor costs and increase of fee income.Coverage of sta expenses illustrates the e ciency of the employee working time as a result of their activity (Monea, 2011).
e postulate that impairment on loan losses shall be covered by net interest income seems natural.Banks earn interest income based on lended assets, and losses on impaired assets shall be, naturally, covered by interest income.
e studies of income and expenses of banks in CEE countries support the premise on interest income and fee and commission income as the most signi cant elements of CEE banks income, and the cost of loan losses and personnel costs as the most signi cant expenses.For example, the analysis of the structure of income and expenses of Roma-nian banks reveals that over 90% of banks' income is found in net interest income and net fee and commission income.On the other side, the larger part of banks' expenses is found in the impairment loss on nancial assets and sta expenses.
e objective of the cross-country analysis of banks which are members of the same banking groups is to identify which of two factors, the residence in the same country or belonging to the same banking group has higher impact on the results of an individual bank's operation measured in terms of three ratios: net fee and commission income compared to net interest income (NFCI/NII), impairment losses on loans compared to net interest income (ILL/NII) and net fee and commission income compared to personnel expenses (NFCI/PE).
Namely, the results of the empirical analysis of competitive conditions in banking systems of the CEE countries over the period 1999-2006 conducted by Delis suggests that the bank's revenue is substantially in uenced by structural and macroeconomic conditions of the country in which the bank operates (Delis, 2008).Previous research on the performance of 515 banks in 16 transition economic during the period 1994-99 indicate that banks' performances di er signi cantly depending on the environment in which they operate (Fries et al., 2002).Also, the ndings of Dietsch and Lozano-Vivas' research on the in uence of environmental determinants on banking e ciency con rm their belief that environmental variables are an important factor in explaining di erences in international banking e ciency (Dietsch & Lozano-Vivas, 2000, p. 1000).Cross-country e ciency studies in the banking industry have generally attracted a lot of attention, but it is di cult to nd studies which compared banks' e ciency for the same group of banks within di erent countries.

METHODOLOGY AND DATA COLLECTION
is paper seeks to examine which of two factors, banking group's policies or the country environment, has higher impact on the su ciency of the bank's net interest income to cover impairment loan losses and net fee and commission income to cover personal expenses.e study is conducted on the panel of banks operating in the CEE region, belonging to four large banking groups, over a two-year period.
e use of panel data is the most suitable tool when the sample comprises of cross-sectional and time-series data.Panel  For the purpose of this paper, data have been gathered from publicly available banks' annual reports and nancial statements for the year 2015 (the list of banks is at the end of this paper).e paper analyses absolute values of three indicators: net fee and commission income compared to net interest income (NFCI/NII), impairment losses on loans compared to net interest income (ILL/NII) and net fee and commission income compared to personnel expenses (NFCI/PE).e indicators used in this text are very simplistic metrics applied on gures from the income statements for 2015 and 2014 therefore, deductions about the bank's e ciency throughout the examined segments using these measures should be interpreted with caution.e research results are presented in very simple way, mainly using graphs and charts in order to enable visual experience.e research also includes a few common simple metrics in the form of "common average", as the rst step needed for higher level overview.

RESULTS AND DISCUSSION
is section analyses distribution of the results /concerning the analysis of relationship between "net fee and commission income and net interest income", "impairment losses on loans and net interest income", and "net fee and commission income and personnel expenses" of 29 banks that are members of the four banking groups and which are a liated in nine CEE countries, as well as their trends over a two-year period.Further details on the presented averages can be achieved by the analysis of the NFCI/NII ratios of individual banks by countries and by bank groups.e results are presented in Chart 1. e data presented in this chart reveal similarities and di erences among individual banks within di erent countries.e general trend based on the analysis by countries is that income of all banks follows similar pattern in most countries.e NFCI/NII ratios of Albanian banks are the lowest ones, within the range of 0.12 and 0.17.e same ratios for Serbian banks are within the range of 0.19 and 0.34, for Croatian banks are within the range of 0.30 and 0.36, for Slovak banks are within the range of 0.30 and 0.36, for Czech banks are within the range of 0.30 and 0.39, for Bulgarian banks are within the range of 0.32 and 0.43, for Romanian banks are within the range of 0.22 and 0.58, for Slovenian banks are within the range of 0.54 and 0.60 and for Hungarian banks are within the range of 0.61 and 0.86.e rank of banks based on the value of NFCI/NII ratio within individual countries varies from country to country.e analysis of these di erences should consider the fact that most of these banks have been transforming themselves into nancial groups by adding subsidiaries, in order to offer additional services (Delis, 2009, p. 6).On the other hand, some banks incorporate additional activities within an individual entity's own activities and present them in their separate nancial statements.ese developments resulted in signi cant modi cations in the balance sheets and income statements of these banks, and must be taken into account when conclusions are made.
Visual presentation of individual banks' NFCI/NII ratio weights in the chart above clearly shows that the weights of indicators for di erent banks within the same country are much more similar to each other than they are within the same group of banks.Taking deposits and placing loans out of these deposits is the core business of banks.Related income arises out of extended loans in the form of interest income and related costs arise out of deposits in the form of interest expenses.Default banking risk relates to losses on non-performing loans, and the cost of this risk should be covered by net interest income.Banks' income statements sometimes report their nancial results following such reporting pattern.Distribution of country average ratios "impairment losses on loans / net interest income" (IIL/NII) varied in the range from 0.12 for Czech to 0.50 for Hungary in 2015, and from 0.09 for Czech and 1.27 for Hungary in 2014.Czech and Slovak banks' loan portfolios were the healthiest banks at that time.On the contrary, Hungarian and Romanian banks experienced the greatest loan losses compared to their net interest income in both years, especially in 2014.Distribution of ILL/NII ratio by countries is presented in Figure 3. e analysis of ILL/NII ratio by banks, as presented in Figure 4, shows that all banks experienced higher impairment loan losses compared to net interest income in 2014 than in 2015.Rai eisen bank and Banca Intesa had a very high ILL/NII ratio in 2014, but signi cantly lower one in 2015.UniCredit ILL/NII ratio was almost the same in both years.Societe Generale bank's ILL/NII ratios were the lowest in both years.
To avoid high variances among the banks during two years, we calculated summarized ILL/NII ratio for two years (the sum of ILL for 2014 and 2015 divided by the sum of NII for both years), by banks and by countries.e analysis of results reveals that Rai eisen bank in Hungary had ultimately the worst experience among all banks (summarized ILL/NII = 1.53), followed by Banca Intesa in Romania (summarized ILL/NII = 1.26), and Banca Intesa in Hungary (summarized ILL/NII = 1.05).Higher weights of "net fee and commission income / personnel expenses" (NFCI/PE) ratios in 2015 than in 2014, in all countries except Romania, is the most signi cant characteristic of (NFCI/PE) trend.e distribution of average NFCI/PE ratio by countries is presented in Figure 6.  e analysis by banks reveals that UniCredit bank is the only one able to nance its personnel expenses from net fee and commission income.Higher value of NFCI/PE ratios in 2015 than in 2014 is the most signi cant characteristic of its trend.e distribution of average NFCI/PE ratio by banks is presented in Figure 7.
A detailed review of NFCI/PE by banks and countries reveals that UniCredit is the bank with most frequently occurring high values of this indicator (over one) within the selected countries ( ve out of seven banks).e next one is Rai eisen (four out of eight banks), then Societe Generale (two out of seven banks) and nally Banca Intesa (one out of seven banks).e overview of NFCI/PE ratios by banks and countries in 2015 is presented in Figure 8.

CONCLUSIONS
e herein conducted research involves two stages.In the rst stage we analyse the absolute values of three indicators: net fee and commission income compared to net interest income (NFCI/NII), impairment losses on loans compared to net interest income (ILL/NII), and net fee and commission income compared to personnel expenses (NFCI/PE) of 29 banks that are members of one of the four banking groups (Banca Intesa, UniCredit, Rai eisen and Societe Generale) within nine Central and Eastern European (CEE) countries.In the second stage, we analyze the di erences and similarities between banks based on their group and country a liation.
e analysis of the NFCI/NII ratios of individual banks by countries and by bank groups in 2014 and 2015 reveals that in most countries all of the banks income follows similar pattern.Moreover, the ranking of banks based on the value of NFCI/NII ratio within individual countries varies from country to country.e di erences are obvious within various countries.e NFCI/NII ratios of Albanian banks are the lowest, followed by Serbian banks, then by Croatian banks, Slovak banks, Czech banks, Bulgarian banks, Romanian banks, Slovenian banks, and nally by Hungarian banks, which have the highest value of NFCI/NII ratios.Distribution of country average ratios "impairment losses on loans / net interest income" (IIL/NII) in 2014 and 2015 varied from country to country.Czech and Slovak banks' loan portfolios were the healthiest banks at that time.On the contrary, Hungarian and Romanian banks experienced the greatest loan losses compared to their net interest income in both years, especially in 2014.In order to avoid high variances among the banks over the period of two years, we calculated summarized ILL/NII ratio for two years by banks and by countries.e analysis of the results reveals that Raiffeisen bank in Hungary had ultimately the worst experience among all banks, followed by Banca Intesa in Romania and Banca Intesa in Hungary.e analysis of the share of personnel expenses nanced by fee and commission income (NFCI/PE ratio) shows that banks' net fee and commission income are usually not sufcient for nancing personal expenses.e NFCI/PE ratio was the lowest in Albania in 2014, it had very similar values for Romanian, Slovenian, Croatian, Serbian, Czech and Slovak banks, and the highest values for Bulgarian and Hungarian banks, which are able to nance their personnel expenses fully by net fee and commission income.
Higher weights of "net fee and commission income / personnel expenses" (NFCI/PE) ratios in 2015 than in 2014, in all countries except for Romania, is the most signi cant characteristic of (NFCI/PE) trend.e nal objective of this paper is to examine which of two factors, the banking group's policies or the country environment, has higher impact on the above results of banks' operations.e overall results reveal that, in spite of strong group management policies, the results of individual banks from the same group vary from country to country, and that there are more similarities among banks from various groups in the same country that among the banks from the same group coming from di erent countries.We may conclude that the country environment has a considerably higher impact on the results of bank's operations than group management policies.
Radovi po pozivu / Invited papers cost/income indicator from 46 in 2011 to 41 in 2015.e values of Banca Intesa and Societe Generale's cost/income ratio signi cantly increased over the same period, while the value of Rai eisen bank's cost/income ratio was almost the same at the beginning and end of the period.

Finiz 2016
Rizici u savremenim uslovima poslovanjaRadovi po pozivu / Invited papers data include banks which operate in the CEE countries, and which belong to four signi cant banking groups: Intesa, Rai eisen, Societe Generale, and UniCredit.e sample contains 29 banks from 9 countries.Out of the total number of banks, 3 are in Albania, 3 in Bulgaria, 4 in Croatia, 3 in the Czech Republic, 3 in Hungary, 4 in Romania, 2 in the Slovak Republic, 3 in Slovenia, and 4 in Serbia.Data for individual banks are annual and they include information obtained from income statements.e herein observed banks account for the vast majority of banking operations in their respective countries.e selection of these banks is based on their relatively large size and vast diversi cation in the region.ese four banking groups are well diversi ed geographically across various CEE markets, even though each of them is present in some markets and absent in others.
RELATIONSHIP BETWEEN "NET FEE AND COMMISSION INCOME" AND "NET INTEREST INCOME"Distribution of country average ratios "net fee and commission income / net interest income" (NFCI/NI) varies by countries from the lowest of 0.15 for Albania to the highest of 0.81 for Hungary in 2015.Based on the ratio values, we can identify three groups of countries.e rst group includes Albania and Serbia, which are the countries with the lowest NFCI/NII ratio.e second group includes Croatia, Slovakia, the Czech Republic and Bulgaria, with a very stable NFCI/NII ratio in the range from 0.33 to 0.36 in 2015.e third group, with high NFCI/NII ratio in the range from 0.44 to 0.81, includes Romania, Slovenia and Hungary.e dynamics of movements of NFCI/NII between 2014 and 2015, measured by annual index, vary from the lowest of 0.98 for the Czech Republic to the highest 1.18 for Slovakia.Based on the index value, we can identify three groups of countries.e rst one includes the Czech Republic and Romania with negative movements of the annual index (0.98 -0.99).e second group includes Serbia, Croatia and Slovenia with slight movements of this index (1.03-1.05), and the third group includes countries with very dynamic movements of the annual index such as Albania, Slovakia, Bulgaria and Hungary (1.09 -1.15).

Figure 1 .
Figure 1.Net fee and commission income compared to net interest income by countries -NFCI/NII ratio

Finiz 2016
Risks in contemporary businessRadovi po pozivu / Invited papers e analysis by bank groups reveals that distribution of ratios NFCI/NII varied by bank groups from 0.33 for Societe Generale to 0.43 for UniCredit in 2015.Moreover, the Uni-Credit had the highest NCFI/NII ratio in the previous year, too.e dynamics of movements of NFCI/NII between 2014 and 2015 by banks, measured by annual index, varied from negative (0.97) for Societe Generale, to moderate (1.01-1.07)for Rai eisen and Uni Credit, up to the highest of 1.15 for Banca Intesa.

Figure 2 .
Figure 2. Net fee and commission income compared to net interest income by selected banks -NFCI/NII ratio

Figure 3 .
Figure 3. Impairment losses on loans compared to net interest income by countries -ILL/NII ratio

Finiz 2016
Risks in contemporary businessRadovi po pozivu / Invited papersRELATIONSHIP BETWEEN "NET FEE AND COMMISSION INCOME" AND "PERSONNEL EXPENSES"With the exception of Bulgaria and Hungary, banks' net fee and commission income are usually not su cient for nancing personnel expenses.e share of personnel expenses nanced by net fee and commission income was the lowest in Albania both in 2014 (55%) and 2015 (61%).e share of personnel expenses in Romanian, Slovenian, Croatian, Serbian, Czech and Slovak banks varied from 85% to 93% in 2014, and from 79% do 97% in 2015.Bulgarian and Hungarian banks are able to nance their personnel expenses fully by net fee and commission income.

Figure 5 .
Figure 5. Impairment losses on loans compared to net interest income by banks and countriessummarized ILL/NII ratio for two years

Figure 8 .
Figure 8. Net fee and commission income compared to personnel expenses (NFCI/PE) by banks and countries in 2015