Customers’ Attitude Towards Diminishing Partnership Home Financing in Islamic Banking

Problem statement: Diminishing partnership or musharakah mutanaqisah contract in home financing is recently introduced by the Islamic banking industry as an alternative of the bay bi’thaman ajil contract which has received many criticism from shariah scholars as mimicking conventional loan. This study aims to examine customers’ attitude upon diminishing partnership home financing and their intention to subscribe into the financing scheme. Approach: This study gauges customers’ perception on the salient features of diminishing partnership home financing from Islamic banks through a survey of 504 respondents comprise officers, managers and academician in Kuala Lumpur. The structural equation modeling was utilized to find out factors that influence customers’ intention for diminishing partnership home financing under the framework of the theory of reasoned action. Results: The main finding indicates that customers perceived the equity sharing features in diminishing partnership home financing mode differ from the conventional loan and other existing Islamic home financing modes. The result from structural equation modeling shows that customers’ intention for diminishing partnership is more influenced by their subjective norms compared to their attitude toward diminishing partnership home financing. Conclusion: The customers have perceived diminishing partnership as more shariah compliant than bay bi’thaman ajil. Afterwards, opinions from people perceived important by customers are giving more influence towards their decision making process.


INTRODUCTION
Similar to food, clothing and security, home is a basic need for everyone. It is a place to dwell in comfort with family. However, home financing in Malaysia is primarily debt based using predetermined rates in conventional and Islamic financing. The conventional home financing is of course not permissible to Muslim as Islam prohibits charging of interest. As an alternative, the Bay Bi'thaman Ajil (BBA) home financing provides a pathway to avoid interest and engaged in transactions involving riba or usury and gharar or uncertainty.
Nonetheless, after 27 years of BBA home financing implementation, there has been growing criticism on its practice due to high pricing, using interest rate as its benchmark, non Shariah compliance and hardship to customers (Meera and Razak, 2005). As BBA is based on fixed rate structure, the cost of payment can be higher than conventional financing as seen during the 1997 Asian financial crisis. One of the major differences between BBA and Diminishing Partnership (DP) home financing is the capitalization of bank's profit upfront in the selling price to customer. In the case of DP, profit is shared between the bank and customer based on their share of ownership. There are also issues on BBA pertaining to Shariah compliance as Islamic bank does not take ownership of the house before selling it to the customer. This violates the legal maxim which requires equivalent counter value ('iwad) for a legitimate sale (Rosly, 2005). According to Dusuki (2008), the current practice is similar to Bay' al-'Inah (sale and buy back) which is a contentious sale as the banks does not take risk and liability in owning the property. This practice is also not acceptable by international scholars (Meera and Razak, 2005).
The DP concept or musharakah mutanakisah was introduced to overcome the criticism on BBA. It is based on the highly successful Islamic Cooperative Housing Corporation (ICHC) in Canada which was established in 1981 out of necessity to avoid the Muslim community from engaging in riba and gharar.
This study aims to gauge customers' perception on DP home financing focusing on its salient attributes and their intention to subscribe to it. The salient attributes are including the concept used, method of computation and pricing, shariah compliant, justice and equality, societal well being and equitable distribution of income.
Literature review: Fishbein and Ajzen (1975) introduces TRA to establish relationship among beliefs, norms and attitudes towards intention to behave. It is a widely studied model from social psychology, which is concerned with the determinants of consciously intended behaviors. Ajzen et al. (1980) further defines attitude as an index of the degree to which a person likes or dislikes an object and a person's attitude toward a behavior is determined by the set of salient beliefs the person holds about performing the behavior. In order to predict attitude from beliefs, there are three steps should be fulfilled. First, there is a need to find out a person's salient beliefs followed by measurement of how the person evaluates the outcome of each salient belief. Secondly, there is a need to measure the belief strength by asking the person to indicate the likelihood that performing a behavior will result in a given outcome. Finally, the outcome is obtained by multiplying the product of each outcome evaluation by the corresponding beliefs strength to predict a person's attitude.
Attitude and Subjective norms form the behavioral intention jointly (Fig. 1). Subjective norm is defined as a person's perception of the most important people to him or her that the person should or should not perform to the behavior and his or her motivation to comply with the specific referents (Ajzen et al., 1980). A person's subjective norm can be predicted by multiplying the product of each normative belief by the person's corresponding motivation to comply.
In term of the tool of analysis, many studies on TRA have been conducted using Structural Equation Modeling (SEM). For example, Choo et al. (2004) focuses the study using TRA on Indian consumers' purchase for new food. Lin and Lee (2004) and Zhikun and Fungfai (2009) utilize the theory to find out the managers' intention to encourage knowledge sharing among employees while Shih and Fang (2004) studies Taiwanese intention to adopt internet banking. Furthermore, Amin et al. (2009) uses SEM in TRA framework to study accounting students' intention to enroll for Islamic accounting courses. Meanwhile, Kim et al. (2009) studies the mobile technology use for shopping in the United States. The results of all the above studies confirmed the relationship between attitude and subjective norm on buyers' intention to the behavioral intention.
Thus, the objectives of this study are; first, to observe the attitude of Malaysian customers toward DP home financing and second, to investigate factors influence customers' intention to subscribe to DP home financing using the TRA framework. The former will be answered through descriptive techniques and the latter through SEM technique. Since limited studies done in the use of SEM for home financing topics particularly Islamic home financing, therefore, in relation with the second objective, the researchers proposed two research questions: RQ1: Can the TRA be used to predict customers' intention to select DP home financing? RQ2: Are the attitude and subjective norm important in explaining the intention to select DP home financing among customers?
Afterwards, four hypotheses are developed to address the above-two research questions namely: H1: Behavioral Beliefs (BB) is significantly affecting Attitude towards DP home financing (ATT). H2: Normative Beliefs (NB) is significantly affecting Subjective Norms (SN). H3: Attitude towards DP home financing (ATT) is significantly affecting Intention to use DP mode home financing (INT). H4: Subjective Norms (SN) is significantly affecting Intention to use DP mode home financing (INT).

MATERIALS AND METHODS
Data collection: Data for this study were collected by means of a self administered survey using purposive sampling technique on 700 respondents comprise officers, managers and academician within the area of Kuala Lumpur, Malaysia. However, due to incomplete information provided by the respondents in the questionnaire, only 504 questionnaires included in the analysis. The instruments to measure the study variables were derived from published literatures. The structure of the questionnaire was adopted from the sample standard questions of Fishbein and Ajzen (1975) used by Taib et al. (2008). Measurement is by means of five points Likert scale, i.e. 1 = Strongly Disagree; 2 = Disagree; 3 = Neutral; 4 = Agree and 5 = Strongly Agree.

Structural equation model: Structural Equation
Model (SEM) is a well-known method to analyze a survey data. It is a statistical methodology that takes a confirmatory approach to the analysis of a structural theory bearing on some phenomenon (Byrne, 2001) and a comprehensive statistical approach to testing hypotheses about relations among observed and latent variables (Hoyle, 1996). Nowadays, SEM is used by social, behavioral and educational scientist as well as biologists, economists, marketing and medical researchers. Raykov and Marcoulides (2000) gives the following characteristics that are actually the main characteristics of SEM: • The models are usually conceived in terms of not directly measurable and possibly not well defined, theoretical or hypothetical construct. For example, anxiety, attitudes, goals, intelligence, motivation, personality, reading and writing abilities, aggression and socio-economic status can be considered representative of such constructs • The models usually take into account potential errors of measurement in all observed variables, in particular in the independent variables. This is achieved by including an error term for each fallible measure, whether it is an explanatory or predicted variable • The models are usually fit to matrices of interrelationship indices-that is, covariance or correlation matrices-between all pairs of observed variables and sometimes also to variable means Structural Equation Modeling (SEM) is basically comprises of two statistical traditions; (i) factor analysis which was developed in the discipline of psychology and psychometrics and (ii) simultaneous equation modeling which was developed in biostatistics and econometrics (Kaplan, 2008). The steps in running SEM are begun with the specification of a model to be estimated. The model tested is a statistical statement about the relations among variables (Hoyle, 1996). Secondly, is to test the goodness-of-fit of the model proposed. Thirdly, test the relationship among the variables through some measurements and last is to interpret the results in relation with the model tested.
A model fit test was carried out to determine whether the model should be accepted or rejected. If the model is accepted, the researchers can proceed to interpret the path coefficients in the model because "significant" path coefficients in poor fit models are not meaningful. AMOS 16 prints numbers of different goodness-of-fit measures and the choice to be used has been a matter of dispute among methodologists. Table 1 presents the model fit measurement and its cut-off value.

RESULTS
Descriptive analysis: The respondents' distribution by gender indicated that 51.4% were males and 48.6% females. Majority of the respondents (51.4%) have an income level less than or equal to RM3, 000. This was followed by 28.8% between RM3, 001 to RM5, 000 and 15.7% between RM5, 001 to RM10, 000. In terms of education level, majority of respondents (60.9 %) were holders of Bachelor's degree (39.1%) and Diploma holders (21.8%) respectively.
Post graduate students with Masters and PhD degrees comprised 20.6%. In terms of working experience, 59.7% respondents have more than 6 years of working whilst the remaining 40.3% below 5 years. These profiles indicated that the respondents are matured and have the level of knowledge and working experience representing house owners and potential house owners needed for the survey. Please refer to Table 2 below for detail breakdown.

Reliability analysis:
Cronbach's Alpha is used to test the reliability of the research instruments. The acceptable value of cronbach's alpha is should be greater than 0.7 (Nunnelly and Bernstein, 1994). The result shows good estimation of the internal consistency reliability as the variables ranged between 0.82 and 0.92 for each latent variable indicated in the model. Table 3 shows the values for every latent variable in the TRA model.  Customers' attitude towards DP home financing: Table 4 illustrates the descriptive analysis on customers' attitude for DP home financing based on the six variables namely concept used (question 1), method of computation and pricing (question 2, 3, 4 and 5), Shariah compliant (question 6 and 7), justice and equality (question 8, 9, 10 and 11), societal well being and equitable distribution of income (question 12, 13 and 14) and customers' preference (question 15). As the answer is formatted using Likert scale i.e. from 1 (strongly disagree) to 5 (strongly agree), the high means and small standard deviations obtained from all the 15 questions indicated that majority of the respondents agreed on the salient features of DP home financing are not similar to the debt-based conventional and BBA home financing. In other words, DP home financing is more shariah compliant and more just compared to BBA scheme and respondents are seriously consider DP for their home financing scheme.  Fig. 1, the dependent variable is intention to use DP home financing (INT).
A model fit test was carried out to determine whether the model should be accepted or rejected. If the model is accepted, the researchers can proceed to interpret the path coefficients in the model. This is because 'significant' path coefficients in a poor fit models is not meaningful. The statistical package of AMOS 16 prints numbers of different goodness-of-fit measures and the choice to be used has been a matter of dispute among methodologists.
Most researchers recommend reporting chi-square as the goodness-of-fit index of the model under the Maximum Likelihood Estimation (MLE). However, chi-square is sensitive to sample size. Schumacker and Lomax (2010) and Hair and Anderson (2010) argue that with large sample size, the chi-square values will be inflated and reports statistically significant, which might erroneously imply a poor data-to-model, fit. Wheaton (1987) advocated not to use chi-square. To overcome the limitations of the chi-squre test, other goodness-of-fit indexes be used as substitutes to the chisquare statistic to assess the model fit such as RMSEA; the baseline fit measures (e.g. NFI, RFI, IFI, TLI and CFI) and GFI under Unweighted Least Square (ULS) method. Accept the goodness-of-fit of the model RMSEA (0.079) <0.08 Accept the goodness-of-fit of the model  As shown in Table 5, p-value for Chi-square (CMIN) is 0.000 which means reject null hypothesis. In other words, CMIN criteria rejects the fit of the model. Baseline comparisons table, however, gives information about NFI, TLI, RFI, IFI and CFI measures which confirmed the acceptance of the goodness-of-fit of the model tested. The values of NFI, TLI, RFI are greater than 0.8 and IFI and CFI values are greater than 0.9. Similarly, RMSEA and GFI values which are less than 0.08 (0.079) and greater than 0.95 (0.989) respectively confirmed the goodness-of-fit of the model tested. Hence, the results confirmed that the model of customers' intention to select DP home financing can be approached by the TRA framework. This result is answering the question addressed in RQ1 and thus confirming that the TRA model is applicable to predict Malaysian customers' intention behavior towards DP home financing.
On the acceptance of the TRA model's goodnessof-fit, the study proceeds to interpret the parameters estimated by SEM. The regression's weights are reported in Table 6. In general, the regression model is significant and meaningful due to the significant pvalue at 1% alpha and the estimated path coefficients are greater than 0.2. These would be meant that all four hypotheses developed are supported.
Variable of BB is significantly and positively (β = 1.175) related to ATT (H1) and also ATT to INT which has β equal to 0.431 (H3). The NB is also significantly and positively (β = 0.634) related to the SN (H2) and also the SN to the INT with β equal to 0.656 (H4). These results are confirming the RQ2 of this study that customers' attitude and subjective norm are able to explain their intention behavior towards the DP home financing. Furthermore, the result indicates that the subjective norm, has stronger influence than the attitude towards the customers' intention behavior on the DP home financing.
As shown in Table 7, the R-square values for the path of NM →SN and the path of BB→ATT are 0.527 and 0.700 respectively, which means that 52.7 and 70% of variation in SN and ATT can be explained by their predictors (NB and BB). The R-square value for the endogenous variable of INT is equal to 0.695 which means that almost 70% of its variation can be explained by its predictors i.e. SN and ATT.

DISCUSSION
The discussion in this study is divided into two parts namely customers' attitude toward DP home financing and the benefits obtained from the analysis of the SEM on the factors influence customers to choose DP home financing. Firstly, the high means obtained from the descriptive results indicated that the customers agreed on the statement of profit sharing concept in DP home financing differs from the prevailing debt based concept in conventional and BBA home financing.
These results provide useful feedback for banks to differentiate their existing BBA home financing product with the newly introduced DP home financing mode. This will overcome the current issues faced by customers which connect between BBA homes financing with conventional home financing as they could not see much difference in the methods of pricing and computing bank's profit. The use of DP, however, can be a better alternative mode which addresses the issue of justice concept and does not cause burden and hardship to customers. This concept allows the customers to gradually own the property and takes ownership upon full payment of the bank's share. The descriptive results show that customers' preference for DP has a high mean score of 3.6 out of 5.
Secondly, the results from the SEM are supporting all the hypotheses developed in this study and automatically confirmed that attitude and subjective norm have influenced on customers' selection for DP home financing. The result implies that customers are more analytical in their selection process for home financing and do not just make their decision based solely upon approval by Shariah board as in BBA home financing. Therefore, bankers can make use the salient features in DP to promote to their existing and new customers through appropriate marketing channels. Likewise, bank can also focus their marketing strategies to promote DP home financing by attracting referent groups such as spouses, parents, scholars and friends to benefit the subjective norm.

Conclusion and suggestions:
This study tries to explore the attitude of Malaysian bank customers towards the newly introduced DP home financing and to determine the factors influence their intention to use that mode of home financing. It finds that customers perceive DP home financing differently from existed BBA home financing and conventional loan. Afterwards, based upon the TRA model, both the attitude towards the DP home financing and the subjective norms are significantly influence customers' intention to subscribe to the DP home financing. However, perceptions of people perceived important by the customers play more role in influencing the intention compared to the attitude.
This study is not without limitations. Further researches are suggested to be directed to; (i) similar research in different area within Malaysia or different countries and (ii) also be suggested to use different sophisticated tools of analysis. Lastly, (iii) advanced researches are also encouraged to analyze the issue of DP from various aspects of banking operation such as banking risks arise due to the implementation of DP and how to develop rental index so that DP home financing can be implemented properly.