The implication of economic freedom on bank efficiency an empirical evidence of vietnamese commercial banksevidence of vietnamese commercial banks

However, the influence of different economic freedom counterparts on efficiency banking sector is not asuniform as economic freedom overall index such as the higher the degree of propert

UNIVERSITY OF ECONOMICS HOCHIMINH CITY

-PHUNG THI LAN NHI

THE IMPLICATIONS OF ECONOMIC FREEDOM ON BANK EFFICIENCY: EMPIRICAL EVIDENCE OF

VIETNAMESE COMMERCIAL BANKS

MASTER THESIS IN ECONOMICS

HoChiMinh City – 2020

UNIVERSITY OF ECONOMICS HOCHIMINH CITY

-PHUNG THI LAN NHI

THE IMPLICATIONS OF ECONOMIC FREEDOM ON BANK EFFICIENCY: EMPIRICAL EVIDENCE OF

VIETNAMESE COMMERCIAL BANKS

Major: Banking and Finance (Research Orientation)

Major code: 8340201 MASTER THESIS IN ECONOMICS

Supervisor: DR LE DAT CHI

HoChiMinh City – 2020

This is to certify that this thesis entitled: “the implication of economic freedom

on bank efficiency: an empirical evidence of Vietnamese commercial banks” which Isubmitted to fulfill the requirements for the degree of master in finance This thesis isonly my original work and due supervision as well as acknowledgment have been made

in the text to material used

Ho Chi Minh City, 6th December

2019Author

Phung Thi Lan Nhi

TABLE OF CONTENTS

ABBREVIATIONS

LIST OF TABLES

LIST OF FIGURES

ABSTRACT

CHAPTER 1: INTRODUCTION 1

1.1 Research objectives 2

1.2 Research questions 2

1.3 Research scope and methods 3

1.4 Research Structure 3

CHAPTER 2: LITERATURE REVIEW 4

2.1 Theoretical Literature 4

2.1.1 Theory of economic freedom 4

2.1.1.1 The concept of economic freedom 4

2.1.1.2 Economic freedom Indicators 5

2.1.2 Theory of bank efficiency 9

2.1.2.1 The definition of Bank efficiency 9

2.1.2.2 Bank efficiency - financial ratios approach 9

2.1.2.3 Bank efficiency - Production Possibility frontier (PPF) approach 11

2.1.3 Economic freedom and bank efficiency 15

2.2 Empirical studies 17

2.2.1 Economic freedom 17

2.2.2 Financial freedom 20

2.2.3 Freedom from corruption 20

2.2.4 Government spending 22

2.2.5 Property rights 22

2.2.6 Business freedom 23

2.3 Control variables 24

2.3.1 Bank size 24

2.3.2 Equity/Assets (Bank capitalization) 25

2.3.3 ROAE (Profitability) 26

CHAPTER 3: DATA AND METHODOLOGY 28

3.1 Research method 28

3.1.1 Data Envelopment Analysis (DEA) 28

3.1.2 The bootstrap DEA method 30

3.2 Model specification 32

3.3 Variables description 34

3.3.1 Variables for DEA 34

3.3.2 Variables for truncated regression 36

3.3.2.1 Dependent variable 36

3.3.2.2 Independent variables 36

3.4 Data source 41

CHAPTER 4: RESULTS AND DISCUSSION 42

4.1 Descriptive statistics 42

4.1.1 Efficiency estimates 42

4.1.2 Truncated regression estimates 44

4.1.3 Vietnam economic freedom overview 46

4.2 Correlation 48

4.3 Result 50

4.3.1 First stage result - Bank efficiency 50

4.3.2 Second stage result - Truncated regression result 53

4.4 Sensitive Test 60

CHAPTER 5: CONCLUSION 65

5.1 Conclusion and policy implication 65

5.2 Limitation 66

REFERENCES

APPENDIX

GCC Gulf Cooperation Council

PCBs Private Commercial Banks

SOBs State Owned Banks

SSA Sub- Saharan African

TE Technical efficiency

Table 3.1: The overview of input and output variables to estimate bank efficiency

scores (DEA) in first stage 35

Table 3.2: The overview of explanatory variables to regress in second stage 40

Table 4.1: A summary statistics of variables used to estimate the bank efficiency (first stage) in Vietnamese commercial banks from 2010 to 2018 42

Table 4.2: A summary statistics of variables used for truncated regression (second stage) to investigate economic freedom effects on bank efficiency during period 2010-2018 44

Table 4.3: Viet Nam economic freedom index 2010-2018 46

Table 4.4: Correlation matrix among variables in truncated regression 49

Table 4.5: Banking efficiency (TE) scores in the period 2009-2018 50

Table 4.6: the distribution of Vietnamese bank Technical Efficiency scores 2010-2018 53

Table 4.7: Second stage result - Truncated regression result 53

Table 4.8: Merger and Acquisition in Viet Nam commercial banks 2012-2015 60

Table 4.9: Sensitive Test 61

Table 4.10: Summarize result between second step regression and sensitivity test 64

LIST OF FIGURES Figure 2.1: Technical efficiency (TE) - Allocative efficiency (AE) - Economic efficiency (EE) 12

Figure 2.2: Input- Oriented approach (IO) 13

Figure 2.3: Output- Oriented approach (OO) 14

Figure 3.1: Variable Returns to scale Model (VRS) and constant Returns to scale Model (CRS) 30

Figure 4.1: Graph of average efficiencies of Viet Nam banks 2010-2018 52

AN EMPIRICAL EVIDENCE OF VIETNAMESE COMMERCIAL BANKS Abstract

The effect of economic freedom on the economy’s well-being has been widelydocumented However, it is absent from the literature of empirical evidence abouteffect of economic freedom on the banking sector This study employs the overalleconomic freedom index and the index’s components which are derived from HeritageFoundation to examine their effect on Vietnamese commercial bank’s effieciency Infirst procedure, we obtain efficiency scores of 39 banks in Viet Nam using DataEnvelopment Analysis (DEA), over the period 2010-2018 with 299 observations Thensecond step, the efficiency scores estimated from DEA method will be regressed oneconomic freedom indexes, applying truncated regression model combined withbootstrapped confidence intervals while controlling for bank specific characteristics.Additionally, we carry out a sensitive analysis using a fractional logit estimator as arobustness check

We find strong evidence supporting that far greater economic freedom positivelyimpacts the efficiency of banks in Vietnamese banking sector However, the influence

of different economic freedom counterparts on efficiency banking sector is not asuniform as economic freedom overall index such as the higher the degree of propertyrights, business freedom and freedom from corruption, the better the bank’sperformance while negative effects of financial freedom on bank efficiency Besides,the empirical findings also show the positive relationship between capitalization andbank efficiency as well as credit risk and bank efficiency

Keywords: Banks, economic freedom, bank efficiency, DEA, truncated regression

bootstrap, Viet Nam

CHAPTER 1: INTRODUCTION

Efficiency is the most important aim of most business including banks, and bankefficiency is a widely discussed topic because of its vital roles in creating a stable andprofitable banking sector then impact on economic growth in each country As aperspective that individual’s liberty to pursue its own economic goals will lead toefficient outcomes is as instinctive as the economics theory itself, we have seen thatmost countries trying to close to economic freedom However, most of studies recentlyjust focus on the relationship between economic freedom (EF) and economic growth,less studies have been carried about effect of economic freedom on banking sector,which is one of the most important financial intermediary playing an important role inproviding funding sources for economic growth (Ferreira,2015) EF plays a vital rolefor the banking sector’s development which categorizes of priority for developingnations and improving efficient banking system in a globalized economy Theoreticallyargued, EF helps to motivate the environment leading to efficiency financial system’sestablishment or being innovative ideas and producing capacities, but the relationshipbetween EF and financial activities still remains vague (Terpilih, 2010)

For example, one of component in economic freedom is financial freedom, andthe effect of financial freedom (FF) in emerging markets of financial markets is noteasily determined as it relies on kind of reform and conditions of financial constrains inthe market (Ağca et al., 2007) As a study from Kose et al., (2003), the influence of FFand integration on banking sector and economic growth is hard to determine because itcould rely on the governance’s quality and institutions They points that the integration

in developing countries is lack of experience, and government could create higheruncertainty shocks to banking sector resulting in lower efficiency Lou et al., (2016)find that the openness with high level leads to shrink of bank performance due to thelack of technologies, skills, the high level of competition and knowledge Higherfreedom and openness also lead to the higher dependence on each others and fragility

of banking sectors in the world such as credit risk, economic and information shock(Anginer and Demirguc-Kunt,2014), which can affect bank efficiency ( BE)

On the other hand, Asharf (2017) finds that freedom increases the bank’sdevelopment levels by decreasing cost and bank credit risk The author also finds thatfreedom helps to remove barriers of trading and stimulate lending diversification whichcreates more bank credit demand and banks’ chances to growth Reducing restrictions

in banking activities due to FF provides more potential chances to improve BE andprofitability (Chen, 2009) Banks tend to have higher profitability, efficiency incountries that have high economic and freedom (Tennant and Sutherland 2014) Astudy from Ahamed (2017) also shows that foreign Banks’s entry following freedompolicies will improve the host country’s banking system because of the “technologyspillover” effects which positively impact on the financial institution’s performance

In Viet Nam, particularly this time, the Association of Southeast Asian Nations(ASEAN) banking system has been preparation for the multilateral- liberalizationbefore of the year 2020 and this is basing on the ASEAN Banking IntegrationFramework (ABIF) in 2014 Therefore, we expect Vietnamese banking system toachieve higher liberalization, freedom and integration level in 2020 This keys on thatthe importance to understand the impact of EF on BE to have appropriate actions forthe aim of maintaining stable and efficient banking system for economic development

in the long term

This study follows this spirit by examining the impact of economic freedom onthe efficiency of Vietnamese commercial banks during the period of 2010 to 2018

- How are the Vietnamese commercial bank’s efficiency scores between 2010 and2018?

- Whether and how economic freedom affects on bank’s efficiency?

1.3 Research scope and methods

The data sample of this study includes 39 commercial banks in Viet Nambetween 2010 and 2018 They are both listed and unlisted banks This study employsquantitative analysis method with using a two-step approach:

- First step: Estimation of efficiency scores by using DEA (Data envelopment

analysis), these efficiency scores are measured by technical efficiency (TE)

- Second step: Bank efficiency scores are regressed against an array of economicfreedom variables and other bank specific factors in truncated regression modelcombined with bootstrapped confidence intervals

The data is unbalanced over 9 years examining: the dependent variable is bank’sefficiency which is measure by Technical efficiency (TE) with DEA method, andindependent variables are bank specific factors and economic freedom indexes derivedfrom The Heritage Foundation 2018

1.4 Research Structure

The study contains five chapters:

Chapter 1: Introduction

Chapter 2: Literature Review

Chapter 3: Data and Methodology

Chapter 4: Result and Discussion

Chapter 5: Conclusion

CHAPTER 2: LITERATURE REVIEW

This chapter provides theories about economic freedom and bank efficiency aswell as previous researches the relationship between them

2.1 Theoretical Literature

2.1.1 Theory of economic freedom

2.1.1.1 The concept of economic freedom

Freedom (Liberalization) thinking and capitalism ideal have been generatedsince classical economics like Adam smith, John Locke and recently Milton Friedman.Since Adam smith, economists have believed that freedom to choose demand andsupply, competitive in business and trade with other countries and ensure the propertyright which is essential components for economic advances (North and Thormas,1973) Moreover, Adam Smith, in the book “ The wealth of Nations”, who hademphasized that the invisible hand role in a free market to help an economycomparatively work and function well , then increasingly wealth of nations MiltonFriedman (1962) said “I believe that freedom society exists because of economicfreedom bringing more efficiency than other solutions in controlling economicactivities”

Hayek foresaw decades ago in the book the road to Serfdom:“the guidingprinciple in any attempt to create a world of free man must be this: a policy of freedomfor the individual is the only truly progressive policy” (F.A.Hayek, 1944) Furthermore,

in “ The constitution of Liberty “ (1960) of Friedrich Hayek, he analyses that economicfreedom should be understood as freedom under government’s law, and freedom doesnot mean that absence of all government actions Therefore, economic freedom is not

an absolute freedom, whereas government actions must have Freedom requires forces,violence and fraudulent to be prevented, except using government forces to make surebest situations with aim for individual efficiency If the government’s coercion isexceed limits, economic freedom will be hurt

In The Heritage Foundation 2014 “A comprehensive view of economic freedomencompasses all liberties and rights of production, distribution, or consumption ofgoods and services The highest forms of economic freedom should provide an absoluteright of property ownership; full freedom of movement for labor, capital and good; and

an absolute absence of coercion or constraint of economic activity beyond that which isnecessary for the protection and maintenance of liberty itself Individuals are free towork, produce, consume, and invest in any way they choose under the even-handedapplication of laws, with their economic freedoms at once both protected and respected

by the state” According to Gwartney and Lawson 2002, EF means the extent to which

a market economy is in place, where the central components are voluntary exchange,free competition and protection of person and property

In conclusion, following The Heritage Foundation (2014): “economic freedom isthe condition in which individual can act with maximum autonomy and minimumobstruction in the pursuit of their economic livelihood and greater prosperity However,the goal of economic freedom is not simply an absence of government coercion orconstraint, but the creation and maintenance of a mutual sense of liberty for all Asindividuals enjoy the blessings of economic freedom, they in turn have a responsibility

to respect the economic rights and freedoms of others within the rule of law.Governments are instituted to ensure basic protections of one citizen over another.Positive economic rights such as property and contracts are given societal as well asindividual defense again the destructive tendencies of others At the end, economicfreedom enhanced and secured by the rule of law, government size, regulatoryefficiency, and market openness, is a vital element of human dignity, enablingindividuals plan and direct their lives in ways that maximize their happiness as they seefit Therefore, economic freedom is the key to achieve the broad-based economicdynamic that ensures lasting growth and increases prosperity for society as a whole”

2.1.1.2 Economic freedom Indicators

There are four indicators of economic freedom: The Fraser Institute, TheHeritage Foundation, Freedom house, and Scully & Slottje (1991) They differ in the

methodswhich they have been constructed, and purposes and the conception ofeconomic freedom embodying The first two Indicators (Fraser Instituteand TheHeritage Foundation) have been using up to now.

 The Fraser Institute: This indicator was produced by Hames Gwartney and Robert

Lawson, and it has been more widely used than any measures of economic freedom

It covers the time period data from 1980 to 2008 It is different with the indexcreated by the Heritage Foundation that is constructed by third party information.The cornerstones of economic freedom include personal choice, voluntary exchange,freedom to compete and security of privately owned property In fact, the indexmeasures : Size of government ( expenditures, taxes, and enterprises); Legalstructure and security of property rights; access to sound money; freedom to tradeinternationally; regulation of credit, labor and business

 Freedom House: First published in 1996 of economic freedom, but then this

publication has been discontinued They defined economic freedom through twodifferent dimensions: lack of state infringements on citizen’s right to exchangegoods and services, and the second one is the state establishmentof the rulesgoverning contracts, property rights and other institutional prerequisites required forthe product of economic affairs The Freedom house includes six indices (freedom tohold property, freedom to earn a living, freedom to operate a business, freedom toinvest one’s earnings, freedom to trade internationally and freedom to participate inthe market economy)

 Scully and Slottje (1991): This was an effort to build the first measures of Freedom

House This data is only available in 1980, including 141 countries and havingfifteen different characteristics: exchange rate system freedom, freedom frommilitary draft, property freedom, movement freedom, information freedom, civilfreedom index of Gastil, classify Gastil – Wright for economy system, printing andpress freedom, broadcast freedom, freedom to travel inside, freedom to traveloutside, peace freedom, working permit freedom, freedom to seek withoutpermission, freedom to hold real estate

 The Heritage Foundation: This index is a series of twelveeconomic measurements

produced by the Heritage Foundation and Wall street Journal This index has anadvantage like The Fraser Institute as continuous data, and this has been constructedsince 1995, including 161 countries and public every year The Index of economicfreedom takes a broad and comprehensive view of a country performance,measuring twelve separate areas of economic freedom Some of the aspects ofeconomic freedom are concerned with a country’s interactions with the rest of theworld such the extent of an economy’s openness and trade However, mostlyassessing the liberty of individuals to use their labor or finances without unduerestraint and government interference Each of the economic freedom plays a vitalrole in developing and sustaining personal and national prosperity Every economicfreedom is individually scored on a scale of 0 to 100 An overall economic freedomscore is a simply average of its scores on the twelve individual freedoms Forpresentational clarity, the twelve economic freedoms are classified into four widelycategories: Rule of law (property rights; freedom from corruption; judicialeffectiveness); Government size (fiscal freedom, government spending; tax);Regulatory efficiency (business freedom, labor freedom, monetary freedom); Marketopenness (trade freedom, investment freedom, financial freedom) Besides theoverall economic freedom index, we have selected these indicators below (in fourcategories above) which are closely related to the banking sector in order to carry onthis study

- Financial Freedom: The index is a measure of banking security as well as

independence from government control Being to access and function well in aformal financial system that ensures the availability of payment and investmentservices to individuals, diversified savings, credit.By extending financingchances and promotingentrepreneurship, an open banking environmentencourages competition to bring the most efficient financial intermediation.Banking and financial regulators by the state that is over the assurance oftransparency and honesty in financial markets can decrease efficiency

- Government spending: The cost of excessive government is a central issue ineconomic freedom Excessive government spending eventually will be financed

by higher taxationor run a great risk of crowding out private economic activity.Even worse, a government’s insulation from market discipline often leads tobureaucracy; lower productivity and inefficiency then undermine economicfreedom

- Property rights: The ability to accumulate private property and wealth is

understood to be a central motivating force for workers and investors in amarket economy The protection of private property requires and judicial systemthat is available to all equally and without discrimination And the transparencyand effectiveness of the judicial system have proven to be key determinants of acountry’s prospects for long term economic growth

- Freedom from corruption: In the perspective of EF, corruption can be expressed

as the integrity’s failure in the economic system or a distortion when specialgroups or individuals are able to take at the expenses of the whole Then acountry imposes numerous burdensome barriers on conducting business, highertransaction costs or bribery

- Business freedom: this index expresses the degree of freedom of entrepreneurs

are able to start businesses or in order to obtain licenses and the ease for closing

a business Obstacles to any of these three activities are able to be deterrents tobusiness and then to job creation

Accordingly to many studies, the best indicator is Heritage Foundation because

it has been mostly based on the policies which governments can control actually(Heckelman, 2000) On the other hand, the index by Fraser institute is mostlyambiguous efforts to measure economic freedom, and updating by each five years will

be hard to consider economy measurement in short term In this study, we considereconomic freedom index provided by Heritage Foundation which is updated by everyyear

2.1.2 Theory of bank efficiency

2.1.2.1 The definition of Bank efficiency

The definition of efficiency in general or particularly BE are common terms inthe literature of economic discipline Productivity and efficiency are normally used foreach other In economics, a firm is considered efficient if it can maximize its output

using a given certain inputs, or it reaches the Pareto optimal (Pareto efficiency as a

measure of social welfare is used by many scholars as their efficiency goal A situation

is optimal only if no individuals can be made better off without making someone elseworse off)

An economic system is said to be more efficient than another (in relative terms)

if it can provide more goods and services for society without using more resources Or

we can consider total factor productivity (TFP) which is productivity estimationincluding all factors of production, mean efficiency can be seemed as productivity and

be measured by the ratio output/input

Productivity (Efficiency) =

Out: such as revenues, profit …

In: costs, fixed assets,

Efficiency is one kind of economy perspective, it has been measured bycomparing between the actual values and optimal value of cost, revenues and profit orany expenses which a firm pursues (Daraio and Simar, 2007)

In general terms, for the definition of efficiency in banking sector, a bank could

be called as efficient only if it is able to produce an expected result with a minimumprovided effort of resources

In particularly, there are two approaches to measure the bank efficiency –Financial ratio and production possibility frontier (PPF)

2.1.2.2 Bank efficiency - financial ratios approach

The approach from financial ratio can be understood as measuring bankefficiency from financial ratios which are counted from financial statements In

banking sector, many researchers measured operating efficiency ratio by countingpercentage between overhead to total assets The lower ratio is the higher efficiency ofthe bank Nowadays, the operating efficiency ratio accordingly to internationalstandard is 0.6, this means any banks which has below this ratio, will be efficientoperating and vice versa Besides, other financial ratios has been preferred such asROA, ROE (Chen and Liao, 2009)

In common, there can be considered into four groups of financial ratios tomeasure bank efficiency: Profitability rates (ROA, ROE, ROS, C/I), margin rates(Netinterest margin), weighted result rates, employment efficiency rates

 Profitability rates

- ROA: (The rate of Return on Assets): is a ratio to measure the ability ofmanagement to utilize the actual financial resources of a bank to createreturns This ratio is widely used to evaluate bank’s performance

- ROE: ( The rate of Return on Equity) is a ratio of financial result to a bank’s owner fund

- ROS: (The rate of Return on Sales): is a ratio of financial result to a bank’s income

- C/I: ( Costs ratio to incomes)

 Margin rates

- Net interest margin: a ratio of interest to assets, and interest spread which can

be understood as a difference between the average interest bearing assets andthe average expenses of interest bearing liabilities

 Weight result rates: reserves balance which is shown as a difference between thebuilding up and dissolution of reserves

 Employment efficiency rates

- Assets/ Number of employees

- Result/ Number of employees

Based on analyzing financial ratios and comparing with other banks and time varying, we conclude about the bank efficiency However, this approach has few

drawbacks, such as financial ratios have to be compared with a consensus standard,while constructing this standard is hard Or different financial ratios has differentconclusion about efficiency, perhaps this bank can be concluded that efficiency usingROA whereas inefficiency in capital (Chen and Liao, 2009)

2.1.2.3 Bank efficiency - Production Possibility frontier (PPF) approach

However, in the production economics, the definitions of efficiency and

productivity are two different concepts While the “productivity” is considered as theentire elements that decide the output’s level achieved with the input provided,

efficiency has a different meaning in comparison with productivity

Efficiency has been approached by the production frontier, which showsoutput’s level which can be reached a peak with the same level of input’s level That isclarified as the ideal relationship between input and output, related to a process to gainthe greatest level of outputs with lowest input’s level And the company producing onthis frontier definitely will be seemed as efficiency There is inefficient producing ifthis is below the frontier and further distance, more inefficiency it is

In 1957 Farrell took the efficiency’s measurement into the higher degree byestablishing the functions of distance between efficiency and producing practicallypoint Academically, he is remembered largely for the celebrated non- parametricmeasure of productive efficiency

There are three kinds of efficiency: Technical efficiency – TE; allocativeefficiency –AE; economic efficiency- EE

TE: The ability of a firm to obtain maximal output from a given set of inputs orientation) or the ability of a firm to minimize input from a given set of outputs(Output-orientation)

(Input-AE: the ability of a firm to use the inputs in optimal proportions, given their prices andcan be defined as an optimal utilization for the cost minimizing combination of inputsEE: AE multiply with TE expressing the overall efficiency of the company – Economicefficiency

EE= TE*AE

Assume a model with 2 inputs (x1 and x2) and 1 output (q) as the figure below

Figure 2.1: Technical efficiency (TE) - Allocative efficiency (AE) - Economic efficiency (EE)

TE = 0Q/0P

AE = 0R/0Q

EE= 0R/0P

The figure above demonstrates three kinds of efficiency with input orientation Where:

SS’: Technical efficiency frontier

AA’: Cost line

The TE measured 0Q/0P must lay between 0 and 1, at the value of one meansthat the firm is fully efficient Reduction in cost if production is at Q’, so at the point Qthe TE firm is efficient but AE is inefficient The economic efficiency EE = TE*AE (orequal to 0R/0P)

Furthermore, the PPF’s theory can be processed with two approaches: Oriented approach and Out-put- Oriented approach The firm with IO is to measure thelowest input’s amount to create a given set of outputs, whereas OO one is in order topredict the maximum output’s level from a provided input’s level

Input- Input- Oriented approach (IO):

Below Figure 2 illustrates a company with x1 and x2 (2 inputs), and 1 output y ,SS’ is a frontier which demonstrates each input’s lowest level that can be used to create

a provided output (Technical efficiency frontier) If a company works on this frontier,

it will be technical efficiency in an input-oriented one because of minimizing theinput’s amount The frontier AA’ (cost line) (which can be built when we already knowthe input-price ratio) defines the optimal input’s level to archive the lowest cost

Figure 2.2: Input- Oriented approach (IO)

 Out-put- Oriented approach(OO) :

Figure 3 demonstrates that the case in which a company has one input (x) andtwo outputs (y1, y2) The ZZ’ curve defines that the highest output level can be reached

by using given input’s level x (Technical efficiency frontier) The company is TE only

if it works on this frontier DD’ is the price information The distance CB’ can beinterpreted in terms of cost reduction

There measures in this case also are bounded by zero and one.

Figure 2.3: Output- Oriented approach (OO)

TE = 0A/0B

AE = 0B/0C

EE= 0A/0C = TE*AE

There are exceptional ones by applying both sides (Beccalli et al., 2006) and notconsensus on the proper results of IO and OO in BE measurement up to now Actually,the foundation of theories about efficiency and BE has been established well long timeago and attracted a variety of academic researches carried on such Berger and Deyoung(1997) with different method of evaluation

Inclusion, efficiency concepts are various and diversified, depending on whichpurpose we can consider as different perspectives However in this study, TechnicalEfficiency (EF) is used to measure BE (TE) of bank that focuses on the ability of bank

to reach outcomes with minimum set of inputs To reach the desired output, we make

an effort to minimize the inputs, and then the technological innovations in the bankingindustry will be reflected on the production frontier That means if a bank operates onthis frontier, this could be considered as technically efficient (TE) And furtherdistance, more inefficient the banks are Besides, we apply efficiency with input-orientation because it seems that IO method is almost preferred to the OO as banks can

focus on managing the input level (costs) ,better than basing on the outputs level.(Dipasha et al., 2012) The measurements and analysis of TE are conducted by DataEnvelopment Analysis (DEA), the next section methodology will discuss further.

2.1.3 Economic freedom and bank efficiency

This section lays out the relevant theories as well as literatures on BE and EFrelated to financial sector performance, while we derive the hypotheses below

In fact, there are no existing models of theory to analyze explicitly the impact of

EF on BE developed yet, while as far as we have known that few related economictheories are likely to have impact on the banking sector:

- Classical economics theory: Classical economics is a broad term that refers to

the dominant school of thought for economics in the 18th and 19th centuries Mostlyconsidering Scottish economist Adam Smith is the progenitor of the classical economic

theory Adam Smith (1776) release of the Wealth of Nations highlights some of the

most prominent developments in classical economics His revelations centeredsurrounding free trade and a concept called the "invisible hand" which served as thetheory for the beginning stages of domestic and international supply and demand Self-regulating democracies and capitalistic market developments form the basis forclassical economics The classical economists were pragmatic liberals (economicliberalism), advocating freedom of the market, though they saw a role for the state inproviding for the common good Smith acknowledged that there were areas where themarket is not the best way to serve the common interest, and he took it as a given thatthe greater proportion of the costs supporting the common good should be borne bythose best able to afford them He warned repeatedly of the dangers of monopoly, andstressed the importance of competition To be related to banking sector, an environment

of banking and finance in which a lowest government’s level interference

and less dependent on central bank as well as financial institutions supervision;especially regulations are limited to enforce contractual obligations and prevent fromfraudulence, this will improve bank efficiency

- Economic theory: Suggestions from Economic theories that EF tends to affect

productive encourages, efforts, and the resource using’s effectiveness Arguments fromeconomic historians that the central components for economic growth are free to supply

or choose resources, trade freedom with others, competing in business and ensureproperty rights (North and Thomas 1976) In addition, the intuitional issues like lackinginvestment or weak systems to protect property right and avoid corruption go ondefeating the total EF and economic potential (North and Thomas, 1976), so it is easy

to link with the relationship between bank performance and economic freedom undereconomic theory This can be indicated that a link EF and BE, in particular, the higherthe level of EF, the bank’s benefit in terms cost advantages and overall efficiency

- Helping hands (Pigou, 1938): However, another theory considers the role of

supervision and less freedom as an appropriate policy such helping hands The helpinghand theory implies that a government’s helping hand will be strong and effective tooffset or even ameliorate the failures of market in case of external power, monopolypower and informational asymmetries In banking perspective, this role of governmentseems as a bank’s official supervision to limit on bank’s activities as well as restrict onbank entries and insurance of bank deposit scheme as proper policies that alleviatefailures of market or help to allocate resources (Barth at al., 2006) This can make bankefficient as much as possible to their roles

- The theory of property rights: Coase (1937); Alchian (1969); Demsetz

(1967) are pioneers for this theory Basing on encourages to direct individuals andorganization’s economic behavior, an investor has ability to take return of mostefficient choices, then he or she can consider and take each alternative choices,compared with net gain among each choice alternative one Finally, choosing a choice

in which can produce the highest benefit for her Obviously, this makes the economyoperating efficiently The incentives behind that direct behavior of investor are based

on property rights of the investor’s decision Property rights theory carries on theimpact of right assignments on investor’s decision by allocating resources andefficiency This analysis is assumed in the terms of efficient market’s neoclassical

model in which this does not have information issues; zero transaction cost and agentbehavior is perfect to maximize the profit from all participants existing in the market(Furubotn & Pejovich, 1972) Besides, each investor who has this right will employsthese rights to gain its wealth By implication in banking sector, such protectionprovides incentive for borrowers and lenders, and decreases protection costs as well asensures property rights, then borrowers, investor and lenders can devote fewerresources with no hesitant.

2001 to 2009 The author uses the DEA method to estimate technical efficiency scores

of banks in the first stage, then second stage using truncated regression modelcombined with bootstrapped to regress the efficiency scores with economic freedomvariables This paper finds that a positive significance effects of economic freedom onbank efficiency in term of cost advantages: higher the level of an economy’s financial,higher the benefits for banks Especially, it tends to be clearer in which countries withmore freedom political systems and governments with higher quality governance.Investigating the effects of low-liberalized policies on banks, Sun and Chang (2011)find that if the openness is lower degree, it can go up the liquidation costs or switchingcosts which may seriously decrease bank performance and get case of bank’s loss.Bank can improve the profit efficiency as a result of the liberalization of Technology

spill-over effect (Chen, 2009) Further argument, banks in a country with higherfreedom’s level tend to have more chances to extent their business to obtain economies

of scale and scope leading to be higher bank efficiency EF brings more chances forbanks to approach credit market and customers from oversea which may improve BE,technology, and develop strategies and globalization experience, thus enhancingoverall bank profitability (Prasad et al., 2003) Baggs and Brander (2006) also show aproof that an association from higher level of freedom with bank credit to privatesectors and higher performance Other studies argue that more openness in the bankingmarkets in terms of increasing foreign penetration, reduces bank margins and improvesthe efficiency of banking system (Claessen et al., 2001) Similarly, Glick et al., (2006)report that the relationship between deregulation and banks usually relates with morebanks taking risk’s measurement, transparency and lending efficiency These positiveeffects could enhance bank’s performance Same idea as that, Flannery (1984)considering the restrictions faced on US commercial banks in establishing more thanone full service office location observes that constraints preventing free entry to thebanking industry may force unit banks to operate with a socially inefficientcombination of inputs

b) Un-supporting evidences:

Of course, others can argue that freedom with excessive degree might makebanks to take greater and more risks then in turn might have a bad performance andeven worse, causing to the recent global crisis There are lots of studies find that EF orliberalization can direct banks to take more risks and destroy bank performance Forexample, Sufian (2014) explores the impact of EF on the efficiency of banks inMalaysia With the two stage approach, computing efficiency score from DEA methodthen bootstrap regression to examine the impact of EF on BE, the author finds that thegreater business freedom tend to reduce the efficiency of bank’s operation The findingsuggests the policy makers and regulators taking action by setting more limits onactivities which banks might take

Baggs and Brander (2006) suggest that liberalization period time, obviouslybanks tend to increase their operation scale then make them take more risk exposurescausing of failure of managing their operations This is associated with the weakgovernance’s corporation impeding bank performance A data with higher frequency ofAebi et al (2012) shows that after the introduction of liberalized policies but lack oftechnology, experiencing and competition advantages in emerging countries, normallybanks in these countries are easy to take more risk and have a low-return period AsViet Nam banking system is still weak in perspective of technology, performance,efficiency, size, and power of competitive, perhaps banks in Viet Nam would havelower performance due to the deregulations and economic freedom Furthermore,Klomp and Haan (2015) find that not proper EF within period of the credit boom canhave relation with higher probability of economic recession Lou et al (2016)documents that emerging economy’s banking systems do not have enough technology,transparency, power of competitive and financial specialty to compete with overseabanks, therefore association with lower efficiency In addition to this, Gulamhussen etal., (2014) argue that banks depend on each other (co-dependence) happening withliberalization among emerging countries with exposure more systemic risk as well asinsolvency risk An evidence of Chortareas et al., (2012) implies that the bankingregulation and supervision’s impact on bank performance seem to depend on theregulation’s type Government intervention is usually mentioned in justification toavoid the monopoly power’s development or bank’s taking risk excessively (Freixasand Santomero, 2004) Evidences suggest that regulation in economy plays an especialrole in the efficiency of bank operation among banking sectors in different countries(Barth et al 2006) Freixas and Jorge (2008) find that the monetary policy’s changesbecause of policies becoming liberalized and more freedom can bring some uncertaintyshocks, then negatively affect on bank financial performance.

Based on these arguments above, the hypothesis can be expressed as follow:

H1: Economic freedom increases bank efficiency

In order to test the hypothesis we also capture the economic freedom counterparts inside this Economic freedom index including Economic freedom overall; Financial freedom, Government spending, Property rights, Freedom from Corruption, Business freedom For the purpose of capturing a country’s boarder environment within economic activities we control other freedom counterparts.

2.2.2 Financial freedom

Chortareas (2013) investigates the dynamics between financial freedomcounterparts of the economic freedom index and bank efficiency in European Union,which finds the higher degree of financial freedom the higher benefits for bank overallefficiency Carrying out properly the financial openness’s process can considerablylower the probability of failure of banks because banks could take advantages frommore chances and diversifying risks, leading to the higher bank performance (Abiad etal., 2008) Ağca et al., (2007) find that FF is positive effect on bank lending andperformance as well because it encourages firms to go up their leverage’s level withmore long-term loans More interestingly, examining Vietnamese banking system’sliberalization policies, Vu and Turnell (2010) find that that FF would decreaseconstraints of regulatory and enhance bank cost efficiency which results in improving

BE A data with 12 Asian economies, Lin et al (2016) find that an increase in thepresence of foreign ownership as under effect of FF and then may enhance bank costefficiency turning to the higher performance

In contrast, Sufian (2014) investigates the relationship between economicfreedom and bank’s efficiency which finds the impact of financial freedom is negative,implying that higher financial freedom reduces the efficiency of banks operating inMalaysian banking sector Viet Nam is in trend of multi- integration and liberalization;

we expect the financial freedom level will have positive effect on bank efficiency

H2: Financial freedom will increase bank’s efficiency

2.2.3 Freedom from corruption

Corruption freedom where corruption is defined as dishonesty or decay, in thecontext of governance, it can be defined as the failure of integrity in the system, a

distortion by which individuals are able to gain personally at the expenses of the whole.Freedom from corruption is expected to promote equitable treatment and greaterregulatory efficiency (Miles et al., 2006)

Sufian and Habibullah (2010) find a significant and positive sign of the relationshipbetween corruption freedom and bank performance Additionally, Demirguc-Kunt andHuizinga (1999) suggest that exist of corruption will affect negatively the efficiency ofbanks This result supports that government should impose more solutions to controlthe corruption, then enhance financial performance of the banking sector A paperexamining the influences of corruption on bank lending in Africa indicates thatcorruption adversely impacts bank lending (Charles I Anaere, 2014) and agent costproblem

In contrast, few studies provide opposite findings to support corruption having apositive impact on bank performance such as Khemaies (2017) finds an importantproof that bank can take advantages from corruption because allowing banks toapproach potential customers under political relationship, as a result bank can increasetheir performance Therefore, based on this assumption, a negative relationshipbetween freedom from corruption and BE is expected A factor from corruption thatredirects banks to correct pre-existent government issues to get benefits from theiradvantage financial trades (Zheng et al., 2013) Faccio (2010) finds that because theadministrative procedures is complicated and consumes a lot of time, corruption couldhelp bank to save time and opportunity costs for bribing money, thereby improving BE.The author also suggests that banks which has a relationship with politicians can avoidtroubles and get advantages to take external and priority funds when using corruptioneffectiveness

Despite the conflicting findings of recent literature on the impact of freedomfrom corruption on bank efficiency, we expect more freedom from corruption willimprove bank lending and thus, improve the bank efficiency in Viet Nam

H3: More freedom from corruption will improve bank efficiency

2.2.4 Government spending

Though government spending is an effective tool to stimulate economic growth,Easterly and Rebelo (1993) point out that productive government spending ontransportation or infrastructure brings higher productivity In fact, Viet Nam is atransitioned economy, which indicates that whether it is over necessity Governmentexpenditure has played the basic stimulation and leverage for economy growth.However, government expenditure comprises a signal of coercion that distorts themarket mechanism’s forces Further arguments, government expenditure plannedproject may “crowd out” private sector which normally have more profitability andefficiency than public sector one Based on Chortareas et al 2013, the governmentspending which is a form of government involvement in the economy tends to benegatively linked to the efficiency in banking sector

H4: Higher level of government spending will reduce bank efficiency.

2.2.5 Property rights

Measurement of property rights index is the extent of protection of propertyrights in a country by law and how the government executes those laws Thecorruption’s level, and individuals and businesses are able to enforce contracts that arealso reflected in this index With regard to banking sector efficiency, the legalframework’s quality relates to enforcement of contracts and how protection of propertyrights plays a crucial role for banks (Chen, 2009) The property rights which reflects towhich collateral and bankruptcy laws protect the rights of borrowers and lenders, andthus facilitate lending Higher property right protection and quality of collateral couldeffectively ensure the bank performance due to lenders and those borrowers could beprotected their rights and deposits in bank and vice versa Thereby, higher efficiency inexecuting contracts could decrease bank’s costs, hence can improve bank efficiency.The bureaucratic control of banks along with lack of healthy competition rendersbanking services highly inefficient and removes accountability mechanism in thesector A study investigates the effect of judicial efficiency on banks’ lending spreads(Luc Laevena and Giovanni Majnoni., 2003) This variable capturing the degree of

property rights protection in a country and gathered from the Index of EconomicFreedom constructed by the Heritage Foundation They find that better enforcement oflegal contracts (property right) is critical to lower the expenses of banks for householdsand firms Finally, it is more efficient for financial intermediation.

However, not all people support for the view securing property rights which areimportant components for efficiency and growth Schmid (2006) shows that a certainlevel of insecurity of rights is sometimes crucially important for the growth of economyand development, while too excessive securing property rights may erode innovationand motivation if a business must fully offset for those effects Another argumentagainst on property rights through private titling is that will have conflicts and existinequality’s level in society generating in this society process, then retarding growthand especially pro-poor growth (Easterly, 2001) Another study points out that propertyrights may reduce bank performance from Andrianova et al., (2008) who find thatproperty right brings more risks to banking system because of incentives for bank tofeel self confident and safer entering to competitive market intensively leading toreduce the efficiency of bank in domestic

Because property right’s protection is very important for banking sectors, weexpect the hypothesis as following:

H5: Higher property rights will be higher bank efficiency in Viet Nam.

2.2.6 Business freedom

Business freedom is a measure of how a business to establish or run a businesswith no interference from the government Regulation and rules with being redundantand burdensome are the most usual barriers to affect business activity Every on itsown, we expect business freedom to improve productivity and expand output,especially with private sector resulting in creating more demand for credit Thus,freedom allowing in business will be conductive for the development of bankingsector Therefore, we also follow some recent literature such as Chortareas et al.,(2013); Sufian and Habibullah (2010), the business freedom will improve bank

efficiency, implying that higher freedom for entrepreneurs to start businesses, will be better performance for banking sector.

H6: Higher business freedom is associated with higher efficiency for Vietnamese banking sector.

2.3 Control variables

In addition to economic freedom factors, this study also examines whether somebank-specified factors have relationship with efficiency or not Based on the existingliterature, this study also gives few expectations on the relationship between eachcontrol variable and bank performance

2.3.1 Bank size

The relationship between size and bank efficiency is vague as many differentviews among studies In perspective of studies that argue a positive relationshipbetween size and efficiency, a report of IMF (2009) gives an idea that economy scalesand less intensive competition for larger bank Large banks have higher performancebecause of their advantages in perspective of economy scales and power of market onloans or deposits (Delis and Staikouras, 2011) An empirical study of Saghi-Zedek(2016) on European banks finds that larger banks usually have better performancebecause of effects from diversification and accessing funding with lower costs, butexpose to default with higher probability De Jonghe (2010) suggests an associationbetween bank size and economy’s scale and scope In the same infrastructure andmanagement with lower costs; large banks have ability to provide more products andservices hence, promoting cost efficiency This is consonant with the finding ofBattaglia and Gallo (2017) who confirm the positive and significant effect of bank size

on performance as the advantages of economies of scale Whereas these studies focus

on the advantage of economies of scale and scope or benefit of diversification,Gulamhussen et al (2014) convinces that larger banks can survive better than smallerbanks because of competitive power in intensively competitive markets An Omran(2007) examining bank in Egypt also suggests that the state-owned banks are normally

largest banks, which have connection with politicians and government supports behindwill perform more efficiency than smaller banks.

In contrast, existing few studies give evidences of opposite relationship betweenbank size and efficiency The study of Kilinc and Neyapti (2012) who point that thatlarge banks will have excessive risks due to self confident themselves with “too big tofail”, which results in exposing them to get chance of bank failure or losses Buck andSchliephake (2013) argues that small size banks are easier in monitoring because oftheir simple operations and lowering conflict’s levels from interests, thus obviouslyleading to the more efficiency Furthermore, Abreu and Gulamhussen (2013) give aproof of evidence in the year of financial crisis 2008 that large banks are easy toinvolve in taking risk activity in order to earn higher returns or exposing themselveswith systematic risk because of risky projects taking, turning to serious losses after that.Karry and Chichti (2013) assess the effect of bank size on technical efficiency andshow the negative relationship between bank size and bank efficiency A study fromMesa et al., (2014) which carries by plotting ratio of efficiency against ten intervals oftotal assets is shown that there is not constant about the relationship between efficiencyand bank size

2.3.2 Equity/Assets (Bank capitalization)

This ratio indicates the capital adequacy in relation with bank risk Obviously,with the higher the equity ratio, the lower bank risks Even though we know that theimportance of capitalization is to explain the financial institution’s performance, there

is still ambiguous relationship For example, Athanasoglou et al., (2008) find that asound capital position is able for banks to have more resilient against unexpected lossesand pursue potential chances in an effectively way Saghi-Zedek (2016) finds thatbanks with higher capitalization achieve less vulnerability because of benefits ofdiversification Mirzaei et al., (2013) report that well – capitalized banks are lessdependent on the cost of external funds, resulting in higher performance It is easy tounderstand that banks with higher equity’s level might reduce capital’s cost, and thenhave a positive effect on bank performance (Molyneux, 1993)

However, as lower capital ratio suggests a relatively risky position, one mightexpect a negative coefficient on this variable (Berger 1995) Meslier et al., (2014) findthat unclear relationship between bank capitalization and performance in Philippinebanking system Dietrich and Wanzenried (2011) find that Switzerland banks does nottransfer in effectively way with higher capital ratio into higher income earnings due toduring the crisis the demand of loan not high.

2.3.3 ROAE (Profitability)

This indicator is to measure the profitability to assets ratios, and the profitability

is very important to evaluate the bank performance or efficiency This measures thebank using its assets and controlling of its expenses to generate a rate of return Asusually, we will think that higher indicator resulting higher efficiency This is realisticand expectable for the result that banks with higher returns can provide better serviceswhich can increase efficiency A study of Hasanul et al., (2017) suggests that a return

on average equity ratio has a considerable effect on bank efficiency in Bangladesh

In contrast, a study finds out a proof of negative relationship between efficiencyand profitability (Hou et al., 2014) Additionally, El-Moussawi and Obeid (2010)suggest that profitability affecting not consensus way in efficiency (positively andnegatively) in the GCC banking sector via the period 2005-2008 Sufian andHabibullah (2009) find that ROA does not have effect on efficiency in Koreancommercial banks through the period 1992-2003 Matthew and Ismail (2006) measurethe technical efficiency and productivity of banks in Malaysia from 1994 to 2000 Theyfind that the efficient banks are characterized by size, not profitability (ROAE).Anotherresult from Iveta Repkova (2015), the author examines determinants of bank efficiency

in the Czech during the period of 2001-2012, showing that ROA has a negative impact

on efficiency

2.3.4 Loans/ Total assets (Credit risk)

Demirguc-Kunt and Huizinigua (1999) find a relationship that higher bank loanratio will be better bank performance because of the theory of trade off - credit risk and

returns Loans products are more cost and profit efficient than other bank assets leading

to high performance (Chen, 2009)

In contrast, there are few studies are different views with the trade off theoryabove: higher risk compensated by higher return A study of Beltratti and Paladino(2015) carrying on 44 countries find that the lower loan’s level stimulates banks toprovide more carefully loans and improves the process of management, which in turnhave a positive impact on bank efficiency The low loan’s lover will reduce the bank’sfunding cost because of reducing bankruptcy costs, and react to asymmetricinformation in better way, then improve bank efficiency (Flamini et al., 2009)

In conclusion, many existing studies carried for considering and determining the relationship between bank specific factors and bank efficiency However, it is worthy to take those factors combined with economic freedom into consideration bank efficiency.

CHAPTER 3: DATA AND METHODOLOGY

This chapter gives descriptions about the research method, variables and data used in this research

3.1 Research method

In order to examine the impact of economic freedom on bank efficiency, using atwo-step approach:

- First step: Data envelopment analysis (DEA) - Estimation of bank technical

efficiency scores by using DEA

- Second step: The bootstrap DEA method – Technical efficiency scores are regressedagainst and an array of EF variables as well as other bank specific factors in truncatedregression with bootstrapped confidence interval

3.1.1 Data Envelopment Analysis (DEA)

Farrell (1957) is seemed as a pioneer for developing the piecewise-linear convexhull approach to create model, though later in 1978 further developed by Charnes,Cooper and Rhodes, who coined the term DEA, applied a mathematical planningmodel “Constant returns to scale” – CCR is for measuring the technical efficiencyfrontier And DEA- The non parametric method is linear programming model TheDEA involves in constructing a non parametric production frontier relied on the realinput-output in a given sample It is worthy that DEA model can be estimated by usingeither CRS or VRS assumptions

The constant Returns to scale Model (CRS)

This method was firstly introduced by Charnes et al., (1978) In order to discussthe DEA method, let assume that the data consists of S inputs and M outputs for each

bank, and xi and yi vectors for the ith bank will be presented as below:

Tec ^ i = min Tec^i, λ { Tec > 0 | y i ≤ ∑ =1 yi λ; and Tec ^ i x i ≥ ∑ =1 xi λ; λ ≥ 0 }, i={1,…, n}banks

Where:

- Input Orientation option

- y is a vector of bank outputs

- x is a vector of bank inputs

- λ is a Nx1 vector of constant

- Tec^i is a technical efficiency score for the ith bank Tec^i =1 indicates that the bank is technical efficient, while Tec^i<1 mean that a bank is inefficient

The programming of linear has to be solved by n times, one by one for every

bank in the whole the sample

The variable Returns to scale Model (VRS)

As the CRS assumption is only appropriate when all banks are operating at anoptimal scale Imperfect competition constraints from government may cause a bank to

be not operating at optimal scale Thus, Banker et al (1984) extend the model byproviding variable returns to scale The use of the CRS specification when all firms arenot operating at the optimal scale results in measures of TE which are confounded byscale efficiencies (SE) In fact, TE in DEA method has been separated into SE (Scaleefficiency) and PE (Pure efficiency) SE means that a bank works at the mostproductive scale in optimum An inappropriate size (too large and too small) may causeinefficiency SE has DRS (Decrease to scales) and IRS (Increase to scales) DRS meansthat a bank which is too large can not reach advantages from that size PE (PureEfficiency) is only a ratio of efficiency not considering of scale The use of the VRSspecification permits the calculation of TE devoid of these SE effects Actually, VRSmodel is essential from the CRS with an additional constrain added to the linearprogramming problem:

 Variable Returns to Scale (VRS): constraint ∑ =1 = 1 ( convexity)

In this method, no functional form for data is required, and allowing aproduction frontier reflected in the weights of the input and outputs can be calculated.Besides, this is based on banks, after setting the production frontier; the result willshow the rank from the most efficiency scores to the least efficiency ones

Figure 3.1: Variable Returns to scale Model (VRS) and constant Returns to

scale Model (CRS)

(Source: Coelli et al., 2005)

However, with the purpose of this study, we employ VRS assumption with oriented to measure the efficiency scores because the CRS assumption is only validity when all banks in the sample have scale at the optimal level (Banker et al 1984) Moreover, advance of technology and changes of regulation may have different effects through banks with different sizes, but VRS assumption allows bank models at the entire levels of technology (Assaf et al., 2011) The reason why this study we aim to use

input-TE as total bank efficiency is to eliminate price changes and simply show bank operation efficiency As AE is difficult to have full price information, and VRS allows

us to count only TE to avoid SE effect instead of counting PE.

3.1.2 The bootstrap DEA method

Simar and Wilson (1999) argue that in the statistical sense DEA method generatesefficiency scores which significantly depend on each other Then, employing the DEAscores for the second step regression as the nạve studies carried might violate the basicmodel assumption required by regression models Conventional approaches to

inference are invalid due to complicated and unknown serial correlation amongestimated efficiencies Furthermore, the DEA efficiency scores are assumed that is anefficiency index in a relative term, not absolute at all To solve this problem, Simar andWilson (1998) suggests a method of bootstrap procedures (Data generating process),which are able to be harmonious inference for the second step regression models Thisbootstrap solution is based on the method of re-sampling, which means that assignoriginal data accordingly to quantities for the interest Recently, in 2007 they furtherdevelop their method to consider the effect of environmental variables on the efficiencyregression The model is present below:

Tec^i = Ei β + εi

Where as :

- Tec^i is TE scores estimated from DEA method of bank N at the time t

- Ei is an environmental variable vector explaining for efficiencies among banks

- β is a parameter’s vector

- εi is statistical noise

The most worthy note in Simar and Wilson (2007) that they solved thecorrelation issues and efficiency scores dependent on each others which might affectregression in nạve regression models Because that procedure can produce biascorrected estimates of Tec^i, thus valid estimates of the parameters in the secondregression model The procedure has a single and double bootstrap (respectively withalgorithm#1 and algorithm #2), in order to carry this purpose study, we apply analgorithm #1 with 2000 bootstrap replication as recommendation The process ofalgorithm #1 can be summarized below:

(1) Calculating TE scores Tec^i for each bank applying the linear programing DEA with input- oriented

(2) Using the method of maximum likelihood in order to estimate the truncated regression of Tec^i on Ei so that estimate β^ of β as well as estimating σ^ε of σε

(3) For every bank i = {1,…n } keep repeating the further steps from 1 to 3 below

with 2000 times to set a bootstrap estimating of Tec^*i,b =1, ,2000

3.1 From the N(0; σ^2ε) distribution and right truncation ( 1- β^Ei ) for drawing εi

3.2 Computing Tec^*i,= β^Ei + εi

3.3 Use Maximum likelihood again to estimate the truncated regression

of Tec^*i, on Ei, yielding estimate (β^*; σ^ε*)(4) Use bootstrap values (β^*; σ^ε*) and original values (β^; σ^ε) to

construct estimated confidence interval

3.2 Model specification

Truncated regression bootstrap model

Estimation of following model: (based on the model of chortareas et al.,2013)

EFF k,t = α+ β1H t +β2B k,t +β3YEAR t +ε k,t (1)

Where:

“k” is individual bank, and “t” represents for time period

EFF: bank technical efficiency scores, used DEA method to measure (3.1.1), and bounded between 0 and 1

H is a vector of economic freedom indicators which derived from Heritage Foundation

Ht = (INDEX, FINFREE, GOVERNINDEX, PROPERTY, CORRFREE, BUSINESS)INDEX: Economic freedom index

FINFREE: Financial freedom index

GOVERNINDEX: Government spending index

PROPERTY: Property right index

CORRFREE: Freedom from corruption index

BUSINESS: Business freedom index

Bk is a vector of bank specific characteristics for every single bank

Bk,t= (EQASk , ROAEk , LNTAk , CRk )