Efficiency of Banks in Croatia

(1)

Efficiency of Banks in Croatia

Marko Novak

^*

, Su-Ying Hsu

^**

* PhD student, Southern Taiwan University of Science and Technology, No. 1, Nan-Tai Street, Yongkang Dist., Tainan City 710, Taiwan

e-mail: [email protected]

** Professor, Southern Taiwan University of Science and Technology, No. 1, Nan-Tai Street, Yongkang Dist., Tainan City 710, Taiwan

e-mail: [email protected]

Abstract

The main focus of this research was to investigate the efficiency of banks in Croatia. After 1990, the Croatian bank market faced a transition from a socialist economy to a capitalist one. During the initial phase of capitalism, the number of banks in Croatia, increased rapidly. Later, because of bank consolidation, the number of banks started to decline gradually. Through the years, the bank market faced many challenges, therefore, the authors wanted to research on the last five years period after Croatia joined the European Union (EU). The data used were from banks’ financial statements from the years 2014 to 2018. The authors used MaxDEA software to measure the data envelopment analysis (DEA) following the Charnes, Cooper, and Rhodes (CCR) research method. The authors examined the interest and non-interest revenues and expenditures obtained from the financial statements. This research employed the input-output analysis, which showed the trends in the Croatian banking system after Croatia joined EU in July 2013. These trends revealed that the efficiency level of large – and medium- sized banks increased from the year 2014 to 2018. Since these banks are mostly foreign-owned it can be said that their efficiency increased after Croatia joined the EU.

Keywords:Bank Efficiency; CCR; Data Envelopment Analysis.

JEL Classification: G21; C67; C88.

Introduction

For the last thirty years, Croatia, like many other post socialistic societies, has been dealing with the transition from socialism to capitalism. In the socialist regime, all the banks were government-owned. However, after the beginning of capitalism, banks began to be mostly private owned and because of that there were more banks in the market that were foreign owned in addition to the private owned banks. With the influx of new banks in the market that are profit-driven, the managers tend to work more in order to obtain good financial results. In the capitalism and open market era, there is excessive competition causing the bank managers to take more risk. However, with higher risks come higher chances of financial instability. One of the most important indicators of financial stability is profitability. In order to obtain profitable results, managers must focus on the indicators of operational performance. Efficiency is regarded as one of the indicators and that is why it is important for the measurement of potential bank failures.

(2)

Financial institutions use financial data to measure efficiency. Financial data is the backbone for any decision that managers make to derive high quality results for the shareholders. Efficiency is one of the most significant aspects of bank profitability, which is crucial for sustainability.

Since the beginning of the century, there have been positive shifts in technologies that were felt unanimously. In the everyday market competition, there are no limits for such shifts; therefore, it is important to measure performances, efficiency being one such parameter for measure.

Companies are comparing themselves in order to achieve better business results. They can learn from each-others’ mistakes and also position themselves on the market in order to get higher profit and sustenance in the market.

In this research, the authors used the DEA approach to analyse the efficiency of the banks in Croatia from the year 2014 to 2018, that is, the period after Croatia joined the EU in July 2013.

Literature Review

This section is divided into two parts. The first part includes four studies conducted on the efficiency of the Croatian banks, and the second part includes four other studies related to the efficiency of the banks in other countries.

Davidovic et al. (2019) researched about the efficiency dynamics of the Croatian banks from the year 2006 to 2015. The authors used the intermediation approach, with interest and non-interest expenses and revenues as the input and output variables, respectively. In addition, a variable returns to scale BCC-DEA output-oriented model was implemented. The research focused on the fact that Croatia joined the EU and how efficiency trends were affected during the global financial crisis. The crisis had a negative effect on efficiency, with the overall efficiency dropping by about 3%. Contrarily, after Croatia joined the EU, the efficiency score increased by about 45%. The authors also found that market leaders are more efficient than other competitors. They also found that the largest banks and the state-owned banks are permanently more efficient than the privately owned banks.

Jemric and Vujcic (2002), studied the efficiency of the banks in Croatia from the year 1995 to 2000. They used the DEA approach and discovered that the most efficient banks are foreign- owned and that the new banks are more efficient than the old ones. Further, smaller banks are globally efficient, but when the variable to returns to scale is considered, larger banks showed efficiency as well. Moreover, the authors also found equalisation in terms of average efficiency in the Croatian banking market, both between and within the peer groups of the banks.

Kraft and Tirtiroglu (1998), investigated the initial phase of the Croatian banking system. Since the start of the transition period in 1990 the number of banks in Croatia more than doubled. The authors used the stochastic-cost frontier methodology and employed data from the years 1994 and 1995. They estimated the X-efficiency and the scale-efficiencies for state and the private banks. New banks were X-inefficient and scale-inefficient compared to the old private or state banks, but they were more profitable.

The main idea of Jurcevic and Mihelja Zaja (2013) were to research the measurement results of the banks and insurance companies using DEA and accounting indicators. The data used were from before and after the global financial crisis. Authors researched about the banking and insurance industry. The results indicated that DEA efficiency scores were the lowest in the years 2007 and 2008 for the insurance industry and the banks, respectively.

Diallo (2016) stated that most of the research studies analysed the country’s financial development by examining the level of private credit issued. However, few researchers used bank efficiency as a measure of development in the financial sector. The main purpose was to research on the external financing of the industries during the financial crisis and the role of bank efficiency in it. DEA was used to measure the efficiency of major banks in different

(3)

countries. The results showed that bank efficiency relaxed constraints and increased the growth rate of financially dependent industries during the crisis. Further, the importance of bank efficiency in mitigating the negative effects of the financial crisis and on the growth level of industries that are mostly dependent on external finance was revealed.

Hassan and Jreisat (2016) researched on the efficiency measures of Egyptian banks and their determining factors, using DEA. Authors were using Data Envelopment Analysis (DEA) to found out the efficiency levels of banks in Egypt. Later, they performed a detailed analysis of the banks by dividing them, first, into large, medium, and small-sized banks; and then into foreign-owned and domestically-owned banks. The number of banks covered was 14 from the years 1997 to 2013. The research results showed that medium-sized banks were the most efficient, followed by the foreign-owned banks.

Sharma, et al., 2015 studied the level and determinants of bank efficiency in the Pacific Islands.

The authors used DEA to show that overall efficiency of the banks is lower in the Pacific Islands than in Australia, which is the home country of major banks. Dynamic generalised method of moments (GMM) and panel data results disclosed that only bank credit and personal expenses have a positive effect on efficiency and not the other bank-specific and macroeconomic factors. This research was helpful in improving the banking system in Fiji, Papua New Guinea, Samoa, Tonga, Solomon Islands, and Vanuatu.

Tsolas and Charles (2015) utilized DEA for measuring the efficiency of Greek banks. The authors were motivated by the concerns arising from the global financial crisis. The cornerstone of this research was that the regulatory capital guidelines on loan loss reserves generated dysfunctional outcomes and that the bonds issued by the Greek government had an important impact on the banks’ portfolio with high risk effects. The purpose of this scientific paper was to incorporate risk and show how the efficiency profile of the Greek banking industry was affected. Efficiency was measured using DEA in which financial risk was proxied by credit risk provisions and investments of the banks in the Private Sector Involvement (PSI). The results of this probabilistic DEA model were derived through a Monte Carlo simulation.

Methodology

DEA is a methodology, used for analysing how managers perform in a company and measure the relative efficiency of productive units. Every DEA has multiple inputs and outputs. The managers usually find a bank that serves as a benchmark and then measure the inefficiencies in the input combination or the slack variables of other banks relative to the benchmark. Based on the empirical data on chosen inputs and outputs of a number of entities, called the decision making units (DMUs), DEA is a non-parametric deterministic methodology used for shaping the relatively efficient production frontier. It is also an alternative to the regression analysis, where a single estimated regression equation is applied to each observation vector. On the other hand, DEA examines each vector or DMU separately.

The biggest advantage of DEA is that it does not require the analytical form of a production function. Based solely on the observed data, it constructs the best practice production function.

The biggest disadvantage of DEA is that the frontier is highly sensitive to extreme observations and measurement errors. Hence, it assumes that random errors do not exist and all deviations from the frontier indicate inefficiency. There are many DEA models. In this research, the authors used the CCR (Charnes and Cooper, Rhodes 1984) model. Charnes, Cooper, and Rhodes made a measure of efficiency for each DMU that was developed from the ratio that represents the maximum of the weighted outputs to weighted inputs. It is important that the weights of the ratio are determined such that similar ratios for every DMU must be less than or equal to unity; hence, multiple inputs and outputs are downsized to a single input and output, respectively, without requiring pre-assigned weights. Thus, a single input-output combination is

(4)

an efficiency measure of a weight function. The following model calculates the efficiency measure for a DMU using a mathematical programming method.

max _, , ∑ / ∑

This equation is subject to:

∑ / ∑ 1, 1, … ,

0, 1,2, … ,

represents the amount of input of the ith type of the j in DMU ( >0, i=1,2,…,m, j=1,2,…,n) and is the amount of output of the for DMU ( >0, r=1,2,…,s, j=1,2,…,n) The weights to find out are from variables and by the programming problem.

Because there is an infinite number of solutions if ( ^∗, ^∗) is optimal, then for each positive scalar a ( ^∗, ^∗) is also optimal. The Charnes-Cooper transformation (1962), one can select a representative solution (u,v) for which:

∑ 1

To solve a linear programming problem that is equivalent to the linear fractional programming problem, the denominator in the above efficiency measure is set to equal one and according to that the transformed linear problem for can be described:

max

∑ ∑ 0, 1,2, … ,

∑ 1

0, 1,2, … , 0, 1,2, …,

Results and Discussion

In this section, the efficiency scores of the Croatian banks from the year 2014 to 2018 are described. The years are chosen such that the trends after Croatia joined the EU in July 2013 can be analysed; and the bank efficiency data and described results are presented. To understand an efficient bank, the authors described it with an efficiency score close to or equal to 1. If the score is closer to or equal to 1, the bank is efficient.

(1)

(2)

(3)

(4)

(5) (6)

(5)

Table 1. Efficiency of the Croatian banks in 2014

2014 CCR

DMU Efficiency Score Vaba Banka Varazdin 1.000000

Tesla Stedna Banka 1.000000

Croatia Banka 1.000000

Hrvatska Postanska Banka 1.000000

Primorska Banka 1.000000

Jadranska Banka 0.918074

Kreditna Banka Zagreb 0.880776

Partner Banka 0.815428

Hypo Alpe-Adria-Bank 0.793976

Veneto Banka 0.783540

Karlovacka Banka 0.768671 Samoborska Banka 0.753643 Zagrebacka Banka 0.753573

Kentbank 0.745131

Imex Banka 0.720919

Stedbanka 0.716998

Podravska Banka 0.683428

Slatinska Banka 0.664348

Erste&Steiermarkische Bank 0.632923

BKS Bank 0.609784

Istarska Kreditna Banka Umag 0.598828

Splitska Banka 0.591282

Privredna Banka Zagreb 0.580105 Banka Splitsko-Dalmatinska 0.574848

Sberbank 0.573024 OTP Banka Hrvatska 0.562110

Raiffeisenbank Austria 0.542660 Source: Authors calculation.

In the year 2014, there were 27 banks in Croatia. Only two out of the 27 banks were state- owned. These were Croatia Banka and Hrvatska Postanska Banka. Croatia Banka was a small- sized bank, and Hrvatska Postanska Banka was a medium-sized state-owned bank. The remaining 25 banks were either domestically or foreign-owned private banks, which showed the free flow of foreign capital as part of the investments in Croatia. State-owned banks showed the highest efficiency scores, along with three other small-sized private banks. The medium-sized Raiffeisenbank Austria was the least efficient with an efficiency score of 0.542660. The same procedure was repeated to find out the efficiency scores for the following year.

Table 2. Efficiency of the Croatian banks in 2015

2015 CCR

DMU Efficiency Score Vaba Banka Varazdin 1.000000

Tesla Stedna Banka 1.000000 Hrvatska Postanska Banka 1.000000 Hypo Alpe-Adria-Bank 1.000000

Imex Banka 1.000000

Jadranska Banka 1.000000

(6)

Table 2 (cont.)

Kreditna Banka Zagreb 0.968330 Zagrebacka Banka 0.838361 Primorska Banka 0.833285

Slatinska Banka 0.783422 Karlovacka Banka 0.760900

Banka Kovanica 0.721581

Stedbanka 0.708554 Sberbank 0.691597 Kentbank 0.684490 Podravska Banka 0.677470

Samoborska Banka 0.644263 Erste&Steiermarkische Bank 0.624130 Banka Splitsko-Dalmatinska 0.619857 Raiffeisenbank Austria 0.571556

BKS Bank 0.518914

OTP Banka Hrvatska 0.511552 Privredna Banka Zagreb 0.488719 Source: Authors calculation.

The efficiency scores for the year 2015 showed that the two state-owned banks were again at the top of the efficiency ranking in Croatia. The other banks with the highest efficiency score of 1 were four small-sized private banks, namely, Vaba Banka Varazdin, Tesla Stedna Banka, Imex Banka, and Jadranska Banka. One medium-sized bank, Hypo Alpe-Adra-Bank also received an efficiency score of 1. Privredna Banka Zagreb, one of the largest banks in Croatia, received a very low efficiency score of 0.488719.

Table 3. Efficiency of Croatian banks in 2016

2016 CCR

DMU Efficiency Score OTP Banka Hrvatska 1.000000

Hrvatska Postanska Banka 1.000000

Stedbanka 1.000000 Istarska Kreditna Banka Umag 1.000000

Raiffeisenbank Austria 1.000000 Privredna Banka Zagreb 1.000000 Primorska Banka 1.000000 Zagrebacka Banka 1.000000

Imex Banka 0.906918

Kentbank 0.801557 Erste&Steiermarkische Bank 0.786002

Samoborska Banka 0.773208

Addiko Bank 0.715836

(7)

Table 3 (cont.)

Sberbank 0.711569 Jadranska Banka 0.709563

Slatinska Banka 0.698799

Podravska Banka 0.667628 Karlovacka Banka 0.644356 Kreditna Banka Zagreb 0.624187 Vaba Banka Varazdin 0.546820 Tesla Stedna Banka 0.189520 Source: Authors calculation.

For the year 2016, the results showed that Croatia Banka, the state-owned bank, was no longer efficient, while Hrvatska Postanska Banka, another state-owned bank, was at the top of the ranking. The other large banks had become more efficient, such that the ratio of medium and large-sized banks to the small-sized ones kept on increasing. Further, eight banks received an efficiency score of 1, which included three small-sized banks, three medium-sized banks, and two large banks. Tesla Stedna Banka, which had an efficiency score of 1 in the years 2014 and 2015, receive the least efficiency score in the year 2016, due to the financial difficulties. The three small-sized Croatian banks with the highest efficiency score of 1 were Stedbanka, Istarska Kreditna Banka Umag, and Primorska Banka. In 2015, Privredna Banka Zagreb had the lowest efficiency score of 0.488719 among all the banks, which changed dramatically the following year. For the year 2016 they had perfect efficiency score of 1. The other big bank in Croatia with the perfect efficiency of 1 is Zagrebacka Banka. It is the biggest Croatian bank. There are two other banks with the perfect efficiency score of 1. Both are mid-sized banks, namely, OTP Banka Hrvatska and Raiffeisenbank Austria.

2017 CCR

Hrvatska Postanska Banka 1.000000 Istarska Kreditna Banka Umag 1.000000 Raiffeisenbank Austria 1.000000 Privredna Banka Zagreb 1.000000

J&T Banka 1.000000

Zagrebacka Banka 1.000000

Imex Banka 0.946305

Kentbank 0.810677 Sberbank 0.759619 Erste&Steiermarkische Bank 0.753653

Samoborska Banka 0.702003 Slatinska Banka 0.669256

(8)

Table 4 (cont.) Jadranska Banka 0.650481 Karlovacka Banka 0.642683

Kreditna Banka Zagreb 0.629378

Primorska Banka 0.595444 Podravska Banka 0.550025 Source: Authors calculation.

In the year 2017, the two largest banks in Croatia, Privredna Banka Zagreb and Zagrebacka Banka, had the highest efficiency score of 1. Two small-sized banks, Banka Kovanica and J&T Banka, received an efficiency score of 1. This implies that the number of small banks in Croatia having the highest efficiency score had been declining since 2014. There were three other mid- sized top efficienct banks in Croatia with the efficiency of 1, namely, OTP Banka Hrvatska, Hrvatska Postanska Banka and Raiffeisenbank Austria. In the year 2017 the bank with the lowest efficiency score was Podravska Banka with the efficiency score of 0.550025.

2018 CCR

J&T Banka 1.000000

Raiffeisenbank Austria 1.000000 Banka Kovanica 1.000000 Hrvatska Postanska Banka 0.954363 Zagrebacka Banka 0.950602 Privredna Banka Zagreb 0.909203

Kentbank 0.827312

Imex Banka 0.825928

Sberbank 0.742388

Slatinska Banka 0.689298 Erste&Steiermarkische Bank 0.678122

Agram Banka 0.674767

Jadranska Banka 0.641606 Karlovacka Banka 0.630858 Samoborska Banka 0.626186

Podravska Banka 0.500544 Source: Authors calculation.

(9)

For the year 2018, two large-sized banks were just below the highest efficiency score of 1, which is still very good efficiency result. The top level efficient banks were mostly small banks combined with mid-size level banks. The state-owned Croatia Banka after being at the top of efficiency since 2014, year 2018 ended almost the last one at the bottom with the efficiency score of 0.59. Bank with the lowest efficiency score than Croatia Banka was Podravska Banka for the second year in a row. Banks with the highest efficiency score were three small-sized banks, Istarska Kreditna Banka Umag, J&T Banka, and Banka Kovanica. The other two highly efficient banks were OTP Banka Hrvatska and Raiffeisenbank Austria which were medium- sized.

Conclusion

The main purpose of this study is to estimate the efficiency scores of the Croatian banking industry from the year 2014 to 2018. Those years are the years for which authors could gather the data for the banks and also are the years after Croatia joined the European Union in July 2013. The authors used an output-oriented DEA model (CCR) to determine the efficiency of the banks in Croatia. The variables that have been chosen from the banks’ financial statements for each year were for the years 2014, 2015, 2016, 2017 and 2018. The authors chose interest and non-interest expenses for the outputs and interest and non-interest revenues for the inputs. The software where the data were used to get the final efficiency results was MaxDEA software.

The Croatian bank market was divided into three types, judging by the size: small, medium, and large banks. In addition, they were divided into private and state-owned banks. Private banks were owned by either domestic or foreign owners. The final results showed that from 2014 to 2018, five to eight banks had perfect efficiency scores of 1. The least efficient score during the observed years was around 0.5, except in the year 2016, when Tesla Stedna Banka had an efficiency score of 0.189520, which went bankrupt the following year. Major differences that can be noticed in the research is that big banks in Croatia were not very efficient in the year 2014 and 2015 and they started to be more efficient in years 2016, 2017 and 2018. Other medium-sized banks followed the same path. On an average, efficiency scores in the Croatian banks from the years 2014 to 2018 were on the scale of 0.5 to 1, with the small exception of Tesla Stedna Banka that went bankrupt.

References

1. Charnes, A., Cooper, W. W., and Rhodes, E., 1978. Measuring the efficiency of decision making units. European journal of operational research, 2(6), 429-444.

2. Diallo, B., 2016. Financial dependence and growth during crises: when does bank efficiency really matter?. Economics Bulletin, 36(4), 2491-2505.

3. Hassan, H., and Jreisat, A., 2016. Does bank efficiency matter? A case of Egypt. International Journal of Economics and Financial Issues, 6(2), 473-478.

4. Jemric, I. and Vujcic, B., 2002. Efficiency of Banks in Croatia: A DEA Approach. Comp Econ Stud 44, 169–193 (2002)

5. Jurčević, B., and Žaja, M. M., 2013. Banks and insurance companies efficiency indicators in the period of financial crisis: The case of the Republic of Croatia. Economic research-Ekonomska istraživanja, 26(1), 203-224.

6. Kraft, E., and Tırtıroğlu, D., 1998. Bank efficiency in Croatia: A stochastic-frontier analysis. Journal of comparative economics, 26(2), 282-300.

7. Davidovic, M., Uzelac O., and Zelenovic, V., 2019. Efficiency dynamics of the Croatian banking industry: DEA investigation, Economic Research-Ekonomska Istraživanja, 32:1, 33-49

8. Sharma, P., Gounder, N., and Xiang, D., 2015. Level and determinants of foreign bank efficiency in a pacific island country. Review of Pacific Basin Financial Markets and Policies, 18(01), 1550005.

9. Tsolas, I. E., and Charles, V., 2015. Incorporating risk into bank efficiency: A satisficing DEA approach to assess the Greek banking crisis. Expert Systems with Applications, 42(7), 3491-3500.