Islamic banks and loan loss provisions

Make Islamic Banks Use Loan Loss Provisions To Smooth Their Results?

Abstraction

The intent of this survey is to look into issues associating to income-smoothing pattern in a new context and to analyze the factors which affect loss proviso for loans and investing in Murabaha, Musharka, and Mudarabah for Islamic Bankss.

Muslim Bankss offers a alone environment to prove income smoothing, compared with conventional Bankss runing in many other states, since they adopt the dynamic provisioning policy. Smoothing is supposed to be perfect when the fluctuation coefficient is nil or/and the finding coefficient is equal to the unit. Results show that 29 from 66 Bankss have a fluctuation coefficient less than 0.5 ( 44 % ) and 49 Bankss have a finding coefficient between 0.5 and 1.

Using panel informations econometric techniques, we do n’t happen grounds of income smoothing pattern through LLP. Such commissariats, which sometimes are referred to as dynamic, statistical or countercyclical loan loss commissariats, merit attending from regulators and supervisors as a tool to heighten fiscal stableness.

Introduction

Income smoothing is described as a common signifier of net incomes direction, that is, directors may diminish or increase reported net incomes to cut down volatility ( Wild, Bernstein & A ; Subramanyam, 2001 ) . Income smoothing is the pattern of utilizing accounting techniques to cut down the fluctuations in net income over clip. So, one motive for income smoothing may come from the market ‘s perceptual experience of the peril of the concern being a map of income variableness. Another motive may be incentive wagess to directors based upon income growing.

Since Bankss have considerable flexibleness in finding the size of the one-year proviso for loan losingss, and because this is a non-cash disbursal, it is an first-class tool for income smoothing.

Many surveies investigate income smoothing patterns conducted by industrial and commercial houses. Few of them are devoted to income smoothing in banking industry and are by and large interested in American Bankss to place these patterns while stressing the relationship between commissariats policy and net income. Banks use loan-loss commissariats for income smoothing intents was foremost explored for the US by Schreiner ( 1981 ) and a considerable empirical grounds back uping this hypothesis has been accumulated ( Greenawalt & A ; Sinkey, 1988 ; Scholes, Wilson, & A ; Wolfson,1990 Collins & A ; Shackelford, 1995 ; Bhat, 1996 ; Niswander & A ; Swanson,2000 ; Anandarajan and Hasan, 2003 ; Anandarajan et al. , 2005 ) . Some surveies, nevertheless, fail to happen grounds of income smoothing through loan loss commissariats ( Wetmore & A ; Brick,1994 ; Beatty, Chamberlain, & A ; Magliolo,1995 ; and Ahmed, Takeda, & A ; Thomas, 1999 ) .

Particularly, those related to Islamic Bankss are nonexistent and study assorted consequences. Ismail and Be Lay ( 2004 ) happen grounds about net incomes and capital direction by Malayan Bankss utilizing loan loss proviso over the period ( 1997-1999 ) . Zoubi and Alkhalil ( 2007 ) confirm this determination on a sample composed from 65 conventional and Muslim Bankss runing in GCC. Directors use loan loss commissariats to smooth net incomes over the period ( 2002-2003 ) . However, Ismail et Al. ( 2004 ) show that bank directors of 10 commercial Bankss in Malaysia, which offer the Islamic banking services from 1998 to 2001, did non utilize loan loss proviso to pull off capital and net incomes. They use realized security additions and losingss to pull off net incomes.

This survey seeks to find if income smoothing patterns is used by Islamic Bankss on a big period and sample and so verifying whether or non bank directors may smooth their consequences by pull offing loan loss proviso ( LLP ) for loans and investing in Murabaha, Musharka, and Mudarabah.

Muslim Bankss offers a alone environment to prove income smoothing, compared with conventional Bankss runing in many other states such as US, since they adopt the dynamic provisioning policy. Furthermore, Islamic Bankss should set up an allowance for loss proviso to absorb any future losingss, because they operate under upon rules of Islamic jurisprudence ( the Sharia ) which requires hazard sharing. Our part lies in proving, for this new context, the hypothesis of income smoothing for Muslim Bankss in a big group of states.

This paper is interested in look intoing if differences in loan loss commissariats ordinances and alterations in ordinance over clip within a state impact the inducements of Bankss to smooth their consequences.

In this survey, we investigate bank directors ‘ usage of discretion in gauging loan loss commissariats ( LLP ) to cut down net incomes variableness. Using 66 Islamic bank one-year informations from Bankscope for the period 2002 to 2006, consequences show that 29 from 66 Bankss have a fluctuation coefficient less than 0.5 ( 44 % ) and 49 Bankss have a finding coefficient between 0.5 and 1. Using panel informations econometric techniques, we do non happen grounds of income smoothing through LLP.

The remainder of the paper is organized as follows. Section 2 presents the features of Islamic Bankss ; chiefly the ordinance of loan loss commissariats and militias. Section 3 nowadayss a literature reappraisal of banking income smoothing patterns. Section 4 explains the research design used to prove income smoothing every bit good as the database. We present the consequences of the empirical analysis in subdivision 5. Section 6 concludes the paper.

Slamic Banks Features

Shariaa Principles

Based on Shari’a Torahs and so Islamic Torahs, the investing attack used by Muslim Bankss is different than the 1s used by conventional Bankss. In fact, Islamic Bankss invest their fund jointly with client through chiefly three different methods, Murabah, Musharaka, and Mudarabah while conventional Bankss invest their financess in loans to clients and in securities. Furthermore, Islamic Bankss should run merely on profit/loss sharing agreements ( Ariff 1988 ) , which implies that returns are variable, dependent on bank public presentation and non guaranteed. While in conventional Bankss depositors are like loaners and do non portion in the success of the endeavor beyond having a preset return. Indeed, Islamic Bankss manage their consequences to guarantee better returns than sedimentation histories. In fact, Islamic Bankss are spouses and portion hazard with both depositors and enterprisers. Hence, Islamic Bankss can be viewed as fiscal mediators.

So, the construct of involvement, used by conventional Bankss, is replaced by net income and loss sharing.

Prohibition Of Gharar

Minutess deemed Gharar are prohibited. Gharar denotes changing grades of misrepresentation refering to the monetary value and quality of goods received by a party at the disbursal of the other. Derived functions merchandising e.g. options are considered as holding elements of Gharar.

Trading and covering in derived functions of assorted signifiers is allowed

Equity Financing With Risk To Capital

Available. Enables several parties, including the Islamic Bank to supply equity capital to a undertaking or venture. Losingss are shared on the footing of equity engagement while net incomes are shared on a pre-agreed ratio. Management of the endeavor can be in one of several signifiers depending on whether the funding is through Mudarabah, Musharaka, etc.

Not by and large available through commercial Bankss, but through venture capital companies and investing Bankss which typically take equity bets and direction control of an endeavor for supplying start-up finance.

Balance Between Moral And Material Requirement

The demand to finance physical assets which Bankss normally take ownership of before resale reduces over extension of recognition.

Excessive usage of recognition and debt funding can take to fiscal jobs.

Furthermore, Hassoune ( 2002 ) argues that, governed by the net income sharing rule, an Islamic bank should hold a low volatility of returns. He shows that, on norm, for three out of the six states representing the GCC, Islamic Bankss ‘ profitableness, measured by ROE, was higher than that of conventional commercial Bankss. So the Islamic bank ‘s profitableness is less volatile than that of the conventional 1. Such a smoothing consequence comes from the ability of the Islamic bank to absorb dazes on assets ‘ returns through net income and loss sharing and accordingly do it less vulnerable to the cyclical nature of returns on assets and costs of liabilities. This technique plays the function of a shock absorber, or an insurance against cyclicality in returns, which the conventional bank can non trust on, because it has to pay involvement charges, which are less flexible” . While Muslim Bankss rely, for their support, on high sums of non-profit bearing sedimentations, or not remunerated current histories.

Provisioning Practice

Another cardinal difference between conventional Bankss and the Islamic 1s is their provisioning pattern. In fact, in Islamic Bankss, purveying is more sophisticated than the provisioning used by other conventional Bankss in that it non merely accounts for current particular and general losingss but goes a measure further in taking a forward looking attack to account for expected losingss. This is closer to what is termed in the current regulative literature as ‘dynamic provisioning ‘ .

Cardinal rule in dynamic provisioning is to put aside commissariats against outstanding loans in each accounting period taking into consideration long-term expected loss. Focus on expected instead than existent losingss helps the bank to smooth out the consequence of existent loan losingss on bank ‘s net income.

For all these grounds, we felt it utile to depict the provisioning pattern in Islamic Bankss.

In order to harmonise the patterns across Bankss, the standard Financial Accounting Standard ( Number 11 ) issued by the AAOIFO pertaining to commissariats and contingent liabilities is flexible and imply a grade of uncertainness. Muslim Banks must keep an equal degree of purveying against the damage of assets and job exposures by acknowledging a general and specific proviso. The former, is based on a per centum of the funding portfolio, allows the bank to cover eventual losingss which are non specifically identified. The 2nd is the sum needed to compose the assets down to hard currency tantamount value if this is lower than cost ( FAS 11 ) . Sheikh Nizam Yaquby provides that“The AAOIFI ‘s activities are considered a cardinal basis that underpins Islamic banking activities by maintaining them off from single, personal logical thinking. The corporate personal logical thinking ( ijtihad ) of the AAOIFI is extremely of import in this critical facet of Islamic economic life. Therefore, these criterions deserve rigorous attachment, avoid confusion, misinterpretation, and ambiguity, and to seek lucidity and sound concern activities» .

As stated by Mann and Michael ( 2002 ) , under this attack “bank ‘s income would no longer be measured net of existent losingss, but cyberspace of parts to the expected loss proviso. Actual losingss would be set against the expected loss commissariats, including expected loss commissariats accumulated in past old ages. However, if a bank made a loss that was greater than the accrued dynamic proviso, so it would likely be appropriate for the extra to feed through straight into the net income and loss account.”

Perez et Al. ( 2006 ) advanced that the dynamic proviso produces level loan loss proviso ratios ( i. e. loan loss commissariats over entire loans ) through the economic rhythm. So, this paper is interested in seeing if such a proviso pattern affected the behaviour of Islamic Bankss in footings of income and net incomes direction as Laeven and Majnoni ( 2003 ) indicate.

Literature Review

An extended literature was devoted to analyzing motive to net incomes direction. The ascertained motivations in the context of unregulated sectors may besides be encountered in the banking industry ; they include one ) minimising political costs, two ) minimising funding costs and three ) the maximization of wealth directors ( Cormier, Magnan and Morad ( 1998 ) ; Stolowy and Breton, 2000 ) .

The usage of loan loss proviso as a tool for pull offing net incomes and capital and in some instances, lending to the signaling procedures of house value, has been observed in many industries, including banking ( Wall & A ; Koch, 2000 ) . The intent of this research is to look into the income smoothing patterns through utilizing loan loss commissariats by Islamic Bankss on a big period and sample.

One of the primary aims of net incomes direction is income smoothing, a pattern aimed at cut downing the variableness of net net incomes over clip. The bing literature provides several accounts about why houses smooth net incomes. Barnea, Ronen and Sadan ( 1975 ) and Ronen and Sadan ( 1981 ) see smoothing a signaling device. Trueman and Titman ( 1988 ) attribute income smoothing to comprehend bankruptcy concerns and Rozycki ( 1997 ) to ways to take advantage of the revenue enhancement codification.

For Goel and Thakor ( 2003 ) smoothing is the effect of directors seeking to increase portion monetary values by cut downing the possible loss that stockholders face when they trade for liquidness grounds.

Particularly, income smoothing is the pattern of utilizing accounting techniques to cut down the fluctuations in net income over clip. One motive for income smoothing patterns in Bankss stems from the market stakeholders ( fiscal analysts, investors, bankers, supervising bureaus ) perceptual experience of concern hazard attributed basically to income variableness. Indeed, investors ‘ determinations are normally taken from a set of ratios such as return on equity, return on assets, the capitalisation rate of net incomes and alterations in the consequences. Then, the demand to follow with sensible degrees of these ratios could take Bankss to pull off their consequences ; this is particularly true that high volatility of returns is associated by investors with an increased hazard ( Ma 1988 ) .

Furthermore, Bankss are more capable to changeless attending of authoritiess that their consequences are fluctuating. Attendant additions can be seen as a signal of monopolistic patterns. Significant diminutions may signal a crisis state of affairs and hence encourage regulators to step in. In add-on, Bankss subject to particular supervising by the cardinal Bankss use to pull off net incomes because it contributes to the fiscal stableness of the economic system. Such monitoring involves altering their accounting consequences, considered an of import index in measuring the overall fiscal strength of the sector.

The regulative capital demand under the Basel Accord ( 1988 ) has besides created inducements for bank directors to pull off net incomes. Banks adjust their consequences as a constituent of equity to run into this minimal demand. This can be achieved via pull offing the biggest discretional accumulations in Bankss ‘ disbursal history, the loan loss commissariats ( Kim et Kross, 1998 and Shrieves et Dahl, 2003 ) .

In rule the usage of Loan loss commissariats is in order to set Bankss ‘ loan loss militias to reflect expected hereafter losingss on their loan portfolios. However, bank directors, holding private information sing the default hazard necessary in gauging the loan loss proviso for each twelvemonth, can exert discretion over the timing of commissariats for certain loan losingss ( Wahlen, 1994 ) , hence they use loan loss proviso to smooth income.

Directors may hold added inducement to smooth income in response to the construction of their compensation bundle. Kanagaretnam, Lobo, & A ; Yang ( 2000 ) argue that the degree of current public presentation relation to the industry median is a cardinal determiner of directors ‘ determinations to smooth income. In a ulterior paper, Kanagaretnam, Lobo, & A ; Mathieu, 2003 explore bank conclude that the demand to obtain external funding and the directors ‘ occupation security concerns appear to be a important thrust of income smoothing behaviour.

Work is besides consentaneous on the usage of loan loss commissariats by Bankss as an instrument of net incomes direction given that the chief concern of a bank is allowing credits ( Gonzalez 2008 ; Hasan and Wall 2004, Bhat 1996 ) . The leaders can easy pull strings these commissariats to increase ( lessening ) when the consequence is lower ( higher ) than desired. They are described as specific accumulations and are more suitable to banking and insurance ( Jeanjean 2003 ) .

Empirically, the consequences of old research are controversial sing the usage of this technique. Wall and Koch ( 2000 ) explain these differences through the diverseness of selected samples and the survey periods. This phenomenon is particularly pronounced before the banking reforms of 1990 when the commissariats were included in full among the regulative capital ( Collins et Al. 1995 ; Moyer 1990 ) .

For U.S. Bankss, most surveies confirm its usage. Net incomes direction utilizing commissariats has been validated by several writers: Greenwalt and Sinkey ( 1988 ) on a sample of 106 Bankss between 1976 and 1984, Ma ( 1988 ) on a panel of 900 observations over the period 1980-1984 and Bhat ( 1996 ) on a sample of the 148 largest Bankss during the period 1981-1991. However, U.S. Bankss will besides prosecute more sharply in such patterns since the acceptance of new reforms on capital adequateness that limit the inclusion of commissariats in the solvency ratio to 1.25 % ( Ahmed et al. 1999 ) .

Other surveies have addressed, in different states, the usage of commissariats to pull off the bank ‘s net incomes. Naciri ( 2002 ) shows that the execution of the Basel Accord in Canada had no consequence on public presentation direction as the Canadian Bankss continue to pull off the proviso for dubious debts before and after acceptance ( 1980-1996 ) . In Japan, Shrieves et Dahl ( 2003 ) and Agarwal et Al. ( 2006 ) show this technique in times of fiscal hurt. Spain presents a particular instance, characterized by adhering regulations on purveying which, in rule, go forthing small discretion for directors. Despite these regulative demands, the survey Anandarajan et Al. ( 2003 ) and that of Perez et Al. ( 2006 ) show that Spanish Bankss use militias to pull off their consequences.

In entire and although consequences are assorted, the old literature has shown, for different contexts, the usage of loan loss commissariats by Bankss ‘ directors as a tool to pull off net incomes and so to smooth income. The inquiry so arises to cognize whether this pattern is besides applied in Muslim Bankss.

Zoubi & A ; Khazali ( 2007 ) , utilizing a sample of 65 conventional and Muslim Bankss ( merely 10 are Islamic 1s ) surveies covers merely 2000-2003, back up the income smoothing hypothesis. They indicate that directors of conventional and Muslim Bankss in the GCC part smooth income via loss proviso. In fact, as their ROA before revenue enhancement and loss proviso for a current twelvemonth is higher than a old twelvemonth ‘s ROA, so direction is expected to increase loss proviso for the current twelvemonth. Similarly, Sundararajan ( 2004 ) find, from a sample of 14 Muslim Banks in 8 states, a considerable smoothing of returns paid to gain sharing investing histories, despite broad divergencies in hazard. Ismail and Be Lay ( 2004 ) besides find grounds about net incomes and capital direction by Malayan Bankss utilizing loan loss proviso over the period ( 1997-1999 ) . These consequences are confirmed Shahimi et Al. ( 2005 ) . Using a panel composed of 15 commercial Bankss which provide Islamic banking merchandises and services over a period 1996-2003, they have highlighted that the Islamic Bankss in Malayan banking system do exercise income smoothing through loan loss provisioning every bit good as their conventional opposite numbers.

Muthalib et Al. ( 2005 ) speculate that directors habitually pattern net incomes direction for timeserving grounds ( Watts and Zimmerman, 1978 ) . They argue that, since Muslim directors adhere to Islamic instructions which discourage timeserving behaviours, so they expected to pattern less net incomes direction than non-Muslim 1s. Using informations for 99 Malayan houses from 1980-1996, Muthalib et Al. ( 2005 ) find no statistical grounds of differences between the net incomes direction patterns of the Muslim managed houses and the non-Muslim managed houses listed on the Kuala Lumpur Stock Exchange in Malaysia. However, Ismail et Al. ( 2003 ) , utilizing informations covering 10 commercial Bankss in Malaysia, which offer the Islamic banking services from 1998 to 2001, show that bank directors did non utilize loan loss proviso to pull off capital and net incomes but they used realized security additions and losingss to pull off net incomes.

Therefore, we expect that Islamic Bankss will act the same manner as conventional Bankss in footings of net incomes direction i.e. they smooth their consequences utilizing the loan loss commissariats.

Methodology

Sample Selection

In order to prove the factors that explain the loan loss proviso and to prove the income-smoothing hypothesis, we collected the information from the international database “ Bankscope ” consisted in set uping a homogenous sample of Islamic Bankss, that is 156 Bankss.

Then, to profit from a long skyline we selected Bankss for which fiscal information ( loan loss proviso for loans and investing in Murabaha, Musharka, and Mudarabah ) , are available in three consecutive old ages. We have so eliminated the Bankss that over the sample period had less than three back-to-back old ages of balance sheet observations, in order to command for the consistence and quality of bank coverage. This restraint reduced the sample to 66 Bankss runing in 19 states over the period ( 2002-2006 ) . 65.2 % of Bankss in the sample semen from Bahrain, Iran, United Arab Emirates, Turkey, Kuwait, Sudan and Yemen. So, our sample is reasonably representative because harmonizing to Mazhar M. Islam ( 2003 ) GCC states capture about 13 % of the entire assets of Muslim Bankss in the universe and is chiefly concentrated in Bahrain, Kuwait, and in the UAE. Indeed, the highest spread of the industry has taken topographic point in the GCC states followed by Malaysia, Iran and Sudan.

Research Design

Several methodological analysiss have been used in anterior research to analyze how directors smooth net incomes via loan loss commissariats. Collins et Al. ( 1995 ) used clip series informations while Barth et Al. ( 1990 ) , Hasan and Wall ( 2004 ) used cross-sectional theoretical accounts to analyze the factors which influence loan loss proviso.

The paper sets an accounting and behavioural model from which we derive a reduced-form equation to prove income smoothing patterns through loan loss commissariats ( LLP ) by Muslim Bankss. This attack is available in two phases.

First, we are interested in analyzing the variableness of the object of smoothing ( net income before revenue enhancements and Zakat ) so as to place the group of smoothing Bankss by utilizing a methodological analysis of Baidleman ‘s coefficient ( 1973 ) and Eckel ‘s coefficient ( 1981 ) .Beildleman ‘s coefficient ( called coefficient of finding ) step the correlativity of the same object of smoothing over clip. Beidleman ‘s method assumes a steady growing of the net income over clip. Eckel ‘s coefficients ( called coefficient of fluctuation ) step the variableness of the object of smoothing with its norm over clip. For Eckel ‘s method, the first coefficient of fluctuation is the standard divergence of the alterations in the net income over the sample period for each bank divided by the mean of the net income.

Smoothing is supposed to be perfect when the fluctuation coefficient is nil or/and the finding coefficient is equal to the unit. Despite the assorted techniques available, smoothing can ne’er be practically perfect. So, to choose the smoothing Bankss, we adopt the threshold of 1 to separate between smoothing and non smoothing Bankss for the finding coefficient and 0.5 for the coefficients of fluctuation ( Chalayer and Dumontier, 1995 ) .

Second, we have to analyze whether Muslim Bankss use LLP to smooth their consequences. The econometric theoretical account used in this research is based on the survey of Perez et Al. ( 2006 ) . Using a sample of 142 Spanish Bankss over a period 1986-2002, Perez et Al. ( 2006 ) indicates that Spain offers a alone environment that has developed and started to use from July 2000 a dynamic proviso called statistical proviso. It is the same provisioning policy practiced by Islamic Bankss.

The reappraisal of the anterior literature indicates that there are several factors that explain the determiners of Bankss ‘ loss proviso. Some of them are non-performing loans ( NPL ) , entire loans ( TL ) and GDP growing as variables to command the recognition hazard or the non-discretionary constituent.

The variable NPL measures the dubious assets of bank I in period T, harmonizing to its balance sheet informations ; this variable is normalized by the entire assets of the bank. TL is the ratio of entire loans to entire assets, is a placeholder for the hazard profile of the bank. Coefficients of NPL and TL are expected to be positive. GDP is the rate of growing Gross Domestic Product in twelvemonth t. It is intended to capture the consequence of macroeconomic conditions ( concern rhythm ) on loan loss commissariats, beyond the hazard profile of a peculiar bank. So the coefficient of GDP is expected to be negative

The empirical literature on Bankss normally use the variable EBTP is the net operating income ( net incomes before commissariats and revenue enhancements ) of bank I in period T, normalized by entire assets of the bank, as a placeholder for net incomes direction patterns. Under the inter-temporal income smoothing hypothesis, as suggested by Beatty et Al. ( 1995 ) , Collins et Al. ( 1995 ) and Moyer ( 1990 ) . Ahmed et Al. ( 1999 ) , we would anticipate a positive relationship between net incomes ( before revenue enhancements and commissariats ) and loan loss militias.

Previous surveies control besides consequence of the capital adequateness ratio on the loan loss commissariats. Kim and Kross,1998 and Ahmed et al. , 1998, indicated that, for low capital Bankss, LLPs declined significantly after the new capital adequateness ordinances. The variable Car is the entire capital ratio of the bank at the start of period T ( terminal of period t-1 ) .

The variables SIZE ( measured by the log of entire assets ) and consolidation ( a silent person variable takes 1 if informations comes from accounting statements of amalgamate banking groups and 0 otherwise ) are included in the theoretical account as a control variable. Bigger Bankss are expected to afford larger loss proviso than smaller Bankss. So, a positive relationship between LLP and SIZE is expected. We do non hold any a priori about the expected mark of consolidation.

Once the recognition hazard and control variables were selected, the undermentioned arrested development theoretical account was used to analyze how LLPs are used in net incomes direction:

LLP it: The value of net particular and general loan loss commissariats of bank I in twelvemonth T normalized by the entire assets

TL it: ratio of entire loans normalized by the entire assets

NPL it: Non Performing Loans normalized by the entire assets

EBTP it: Net incomes before revenue enhancements and commissariats normalized by the entire assets

Car it: Capital Adequacy Ratio approximated by entire financess to entire assets

Size it: Logarithm of Total Assets

Con it: Dummy variable takes 1 if informations comes from accounting statements of amalgamate banking groups and 0 otherwise

GDP it: The rate of growing Gross Domestic Product in twelvemonth T

We estimate theoretical accounts ( 1 ) utilizing bank-specific random effects. Our empirical specification follows closely the theoretical accounts used in the literature to prove the income-smoothing hypothesis ( see, for illustration, Greenwald and Sinkey, 1988 ) .

Empirical analysis and consequences

Descriptive Statisticss

Table 1 provides the descriptive statistics for all the variables used in our appraisal sample.

The ratio of loan loss commissariats to entire assets equals 0.54 per centum on norm with a upper limit of 8.9 % per centum. Based on a standard divergence of 0.88 per centum, we advance that there are differences between Bankss in footings of loan losingss purveying patterns. Our consequences are closed to those of Perez & A ; Al. ( 2006 ) who found that the ratio of loan loss commissariats to dawdle entire assets is about 0.65 per centum on norm ( with a standard divergence of 0.88 per centum ) . Indeed, Zoubi et Al. ( 2007 ) , study that conventional and Muslim Bankss in the GCC part make a really low estimation of loss proviso ; they allocate, on norm, merely 1.31 % of their entire loans and investing to the loan loss proviso with a upper limit of 25.68 per cent. On mean NPL to entire assets is 4.23 per centum with a upper limit of 20.98 per centum, which is similar to the degree of 7.53 per cent founded by Zoubi & A ; Al. ( 2007 ) .We discovery besides that Islamic Bankss in our sample estimated their entire credits to be 53.4 per cent of their entire assets. Varing between 0 and 100 % with a standard divergence at 21.33 per centum, this indicates a big scattering in the degree of credits collected by Islamic Bankss. The norm of benefit before revenue enhancements and commissariats to entire assets is 2.29 per centum with a upper limit of 26.11 per centum ( ROA before revenue enhancement and loss proviso reported by Zoubi et Al. ; 2007 is at 2.23 per centum ) . Entire financess are really high, they represent 20.04 times the entire loans and investings for our sample. On norm, the natural logarithm of entire assets is at 13.73 per centum and the standard divergence is 1.66 per centum, which indicates a small divergence of the size from their mean.

Where LLP it: The value of net particular and general loan loss commissariats of bank I in twelvemonth T normalized by the entire assets ; NPL it: Non Performing Loans normalized by the entire assets ; TL it: ratio of entire loans normalized by the entire assets ; EBTP it: Net incomes before revenue enhancements and commissariats normalized by the entire assets ; CAR it: Capital Adequacy Ratio proxied by entire financess to entire assets ; SIZE it: Logarithm of Total Assets.

the correlativity matrix of the arrested development variables. The correlativity coefficients among the independent variables are really low proposing the absence of multicollinearity job among the variables. The correlativity between loan loss commissariats and net income before revenue enhancement and loan loss commissariats is negative, proposing that Bankss do non exert income smoothing on norm. The correlativity between loan loss commissariats and loan growing on the one manus and loan loss commissariats and non executing loans on the other manus are positive corroborating the dynamic proviso pattern by Islamic Bankss. The correlativity between loan loss commissariats and GDP growing is besides negative, proposing a anti-business cyclical behaviour of bank ‘s loan loss provisioning.

Where LLP it: The value of net particular and general loan loss commissariats of bank I in twelvemonth T normalized by the entire assets ; NPL it: Non Performing Loans normalized by the entire assets ; TL it: ratio of entire loans normalized by the entire assets ; EBTP it: Net incomes before revenue enhancements and commissariats normalized by the entire assets ; CAR it: Capital Adequacy Ratio approximated by entire financess to entire assets ; SIZE it: Logarithm of Total Assets ; GDP it: The rate of growing Gross Domestic Product in twelvemonth T ; CONS it: Dummy variable takes 1 if informations comes from accounting statements of amalgamate banking groups and 0 otherwise

Arrested development Consequences

Table 3 shows the figure of drum sander and non-smoother houses that were determined harmonizing to the Beidleman ‘s and Eckel coefficient. Results show that 29 from 66 Bankss have a fluctuation coefficient less than 0.5 and 49 Bankss have a finding coefficient between 0.5 and 1. To get the better of these arbitrary thresholds, we cross the consequences from the fluctuation coefficients with those from the finding coefficients. This allows the designation of 26 smoothing Bankss ( 39,4 % of the sample ) .

This determination confirms that Islamic Bankss manage their consequences to cut down the fluctuations in net income over clip. The tabular array shows that finding coefficients can observe a figure higher for smoother Bankss ( 75 % ) than the Eckel coefficient ( merely 44 % ) . Given the fact that, over the survey period ( 2001-2006 ) , Muslim Bankss have experienced a growing stage and an enlargement of their credits, they are likely to increase their consequences displayed from one twelvemonth to another and non reproducing the same consequence.

Once income smoothing patterns are highlighted by Islamic Bankss, so we investigate if they use Loan loss commissariats to smooth their consequences. Table 4 contains the chief consequences of our econometric probe. It reports four arrested developments utilizing:

-the full sample

– the panel that includes observations for which the finding coefficient is between 0.5 and 1.

– the panel that includes observations for which the fluctuation coefficient is less than 0.5.

– the panel that includes observations for which the finding coefficient is between 0.5 and 1 and the fluctuation coefficient is less than 0.5.

For these four arrested developments we test two specifications ; the first specification includes all the explicative variables defined in our theoretical account equation. In the 2nd specification, we replace the GDP variable by the lagged GDP ( Salas et Saurina, 2002 )

The consequences presented in the paper lead us some decisions.

I ) The income smoothing hypothesis was non supported either in the whole sample or in the sub-samples. The benefits before revenue enhancement and commissariats scaled by entire assets is a non important factor in finding the sum of loss proviso in all specifications. This consequence suggests that Islamic Bankss do non utilize loss proviso to pull off and smooth their consequences. This determination confirms those found by Ismail et Al. ( 2004 ) and Abdul Rahman et Al. ( 2005 ) . They find that houses holding a bulk of Muslim directors will fall back to less net incomes direction than houses that have a bulk of non-Muslim directors. However, this determination is different from those obtained by Zoubi and Khazali ( 2007 ) who report that Islamic Bankss runing in GCC states manage their consequences like conventional Bankss. This consequence could be explained foremost, by the fact that Islam is against self-interest. It encourages its followings to get wealth but prohibits them from geting the wealth utilizing subjugation and out agencies ( Rahman et al. , 2005 ) . Second, dynamic commissariats policy adopted by Muslim Bankss can of course supply stable consequences without any intercession. So, this sort of proviso virtues attending from regulators and supervisors as a tool to heighten fiscal stableness. Finally, the rule of the “Balance Between Moral and Material Requirement” adopted by Muslim Bankss provides the demand to finance physical assets, which Bankss normally take ownership of before resale, and let them to cut down over extension of recognition and hence their profitableness. In contrast, conventional Bankss use overly the recognition and debt funding, which can take to more fiscal hazards jobs.

two ) Significant and robust consequences of the placeholder of the capital adequateness ratio on loan loss proviso were found for all specifications in our arrested development. The negative mark of the CAR ratio indicates that, overall, the lower the capital ratio, the higher the loan loss proviso. This determination supports the fact that Islamic Bankss good capitalized involve in less hazardous activities esteeming net income and loss sharing chief in contrast with conventional Bankss. Banks with high degrees of CARs might be encouraged to ship in riskier activities taking to riskier recognition portfolios. Furthermore, this is consistent with the findings of Moyer ( 1990 ) and Beatty et Al. ( 1995 ) whom provide that loan loss commissariats are used as a mechanism to increase loan loss militias and hence the capital ratio.

three ) All the coefficients of the recognition hazard placeholder variables have the expected mark. The coefficient of non-performing loans over entire assets, NPL, is positive and important. The coefficient of loans over entire assets, TL, is besides positive and important but merely for the whole sample ( first theoretical account ) . So, we can progress that the dynamic provisioning is a pattern followed by Muslim Bankss but this determination should be performed with attention since the variable TL is non important for the sub-sample specifications.

GDP and lagged GDP Growth have mixed consequences and they are non statistically important in all specifications. The concern rhythm does non impact current loan loss proviso determinations of Bankss, after commanding for the remainder of the variables.

four ) The variable CONS which control for the histories consolidation does non hold a important consequence on loan loss commissariats. The log of entire assets is another control variable, mensurating the size of the bank, have a positive and important consequence for all specifications. This determination confirms those found by Zoubi et Al. ( 2007 ) who report that larger Bankss may hold higher degrees of concern and may be expected to hold higher loan loss commissariats than smaller Bankss.

Decision

Policy Deductions

In decision, consequences show that income smoothing practiced by Islamic bank is natural, ensuing from an income bring forthing procedure and non from the deliberate troughs ‘ will to minimise variableness of consequences over clip. This consequence is confirmed in so far as the fluctuation coefficient of the net income before revenue enhancements and Zakat is higher than the investings in Mudaraba, Murabaha, and Musharka of the bank equivalent to his turnover.

Bibliographie

FAS 11: “Specific Provision which is the sum set aside to reflect devaluation of a certain plus i.e. , compose it down to its current hard currency equivalent value. General Provision which is the sum set aside to reflect a possible loss that may happen of current unidentifiable hazards in relation to Total of Assets “Receivables and Investment and Financing” .

Investings in Mudaraba, Murabaha, and Musharka equivalent to entire recognition, net income for Muslim Bankss includes conventional net income before revenue enhancements and Zakat, not executing loans or delinquent.

The distribution of 66 Muslim Bankss by 19 states is: Bahrain ( 11 ) UAE ( 7 ) Turkey ( 6 ) Iran ( 6 ) Sudan ( 5 ) Pakistan ( 5 ) Kuwait ( 4 ) Yemen ( 4 ) Brunei ( 2 ) Qatar ( 2 ) Malaysia ( 2 ) Bangladesh ( 2 ) Arabia saudia ( 2 ) Egypt ( 2 ) Jordan ( 2 ) Russia ( 1 ) Indonesia ( 1 ) Mauritania ( 1 ) Tunisia ( 1 ) .