Faculty of Economics, University Gunadarma
Jl Margonda Raya No. 100 Depok
ABSTRACT
In giving credit, banking financial institutions and non-banking set eligibility standards of a prospective borrower to get a loan. standards and procedures are done to avoid a credit crunch that might happen in the future, such as the debtor could not repay their debts because of one thing or another. Based on this background, researchers formulate problems apasajakah factors that influence the selection of a proper of the debtors get loans
The object of this research is a cooperative that is located in Tasikmalaya KOPPAS HIPPATAS. To determine the factors that influence, can use discriminant analysis.With discriminant analysis can create a model that can clearly show the differences and to classify cases into the current group of debtors, whether or not current in payment of loans.
The results of this analysis, four factors are analyzed to determine the factors that determine the credit worthiness of KOPPAS HIPPATAS consisting of loan size, age of the borrower, mortgage installments owned and provided, there are two of the most significant predictor of influence. Two of these predictors is owned and dependents proposed loan. The amount of loans are the most dominant or significant in differentiating the behavior of borrowers in repaying credit.
Key words: Factors, Lending, KOPPAS HIPPATAS.
INTRODUCTION
In Indonesia's position in micro, small and medium enterprises (SMEs) has long been recognized as a very important business sector, because of various real role in the economy. Starting from the share has in the formation of GDP in 2005 approximately 63.58%, the ability to absorb labor force amounted to 99.45%, or a very large number of units involved ie approximately 99.84% of all existing business units, until the share has a significant in total value of total exports, which reached 18.72%.
Maybe with a very strategic position is the Indonesian macro economic conditions for this to survive and not because of the economic collapse is still felt. With these considerations, as well as the pressure-pressure from various parties to the banking sector in order to channel more credit to the MSME sector, since the beginning of the year 2006 is generally better banking institutions and non-banking, seeks to provide easiness in giving loans but with still did not forget the principle of prudence.
In providing loans, financial institutions set the standards proper of a prospective borrower. Penetapkan standards and procedures are done to avoid a credit crunch that might happen in the future, such as the debtor could not repay their debts because of one thing or another. Usually they select the first customer / credit-worthy borrowers who, for example by conducting surveys to house prospective borrowers. Besides the company is also considering the relevant factors from the debtor, so the company can estimate whether the customer setidaknnya the future will be able to repay their debts or not.
Credit terms derived from Latin (credere) means kepercaan or to believe or trust. Therefore, the rationale credit approval by a financial institution or bank to a person or business entity based on the belief (faith).
10 years in 1998 states that credit is the provision of money or bills are similar, based on the approval or agreement between bank lending and borrowing with another party that requires the borrower to repay the debt after a certain watu term by giving flowers.
The ability and willingness of borrowers repay loans is strongly influenced by external and internal factors that called the C's of Credit. (Siswanto. 2008,73) (Siswanto. 2008.73)
- Character.'s Character, traits, habits debtor (party who owes) is very influential in granting credit.
- Capacity. Capacity is related to the ability of a debtor to repay their loans.
- Capital. By looking at the number of capital owned by the debtor or to see how much capital invested in its business debtor.
- Warranty. Assurance required as a precaution in case the debtor can not recover their loans.
- Economic Conditions. Economic conditions around the dwelling prospective borrowers should also be considered to take into account the economic conditions that will occur in the future.
According Siswanto (2008:74) for the activities of credit analysis, account officer assigned to assess the quality of the credit requests submitted prospective borrowers, analyze internal and external factors above is determined by the following things:
- Total credit to be given.
- The credit period.
- The type and amount of credit guarantees to be provided by prospective borrowers.
- Prospective borrower's reputation and his company the community.
- Prospective debtor relationship with the bank.
Debtors who have dikaregorikan doubtful and loss need to get special attention from the banks, the continuation of action by organizing a rescue (rescue operation). Rescue actions that can be done is as follows (Tjoekam, 1993):
- Rescheduling
This policy relates to the credit period so that relief can be granted are:
- Extend the credit period.
- Extend the repayment period.
- Decrease the amount for each installment.
- Reconditioning
Assistance provided in the form of waivers or modification of terms, among others:
- Capitalized interest
- Postponement of payment of interest
- Lower interest rates
- Interest Exemption
- Pengkonversian kredit jangka pendek menjadi kredit jangka panjang dengan syarat yang lebih ringan.
- Restructuring
Action can be taken within the framework of the restructuring are:
- Additional Credits (Injection / Nursery Operation)
- Additional Equity
- Combination
Is an act that combines some of the action alternatives resheduling, reconditioning and, restructuring.
RESEARCH METHODThe data used in this study the authors obtained by taking the data contained in Market Traders Cooperative "KOPPAS HIPATAS" which addressed at Jl. Residen ArdiwinangunResident Ardiwinangun Home Market I Cikurubuk-Tarlac City. Telp.0265-344798. Telp.0265-344 798.
The sampling method used was random sampling with the number of debtors 141 people. Rescoe in Sekaran (2000) states that the sample size of more than 30 and less than 500 have been sufficient to be used in all studies.
Referring to rescoe opinion, the total number of samples in this study as many as 141 people from the population contained in the KOPPAS HIPPATAS deemed to have been sufficient.
To analyze the data using descriptive analysis to describe the general condition of Cooperatives and discriminant analysis to identify distinguishing between groups. Based on this function, the observation group of unknowns can be determined group. Therefore, discriminant analysis can be used as a method of classification.
DISCUSSION
1.4 Description of Data Debtors
Data used in this study the researchers report in the form of strengthening capital receipts KUKM Koppas Hippatas consisting of debtor data current / noncurrent, sex, age debtors, sex, period, large loans, types of businesses, large installment, and the guarantee given .Such information is necessary for the development of instruments and facilitate the implementation of primary research. To provide a better picture of the data obtained by the researchers, the researchers describe the debtor data through graphs.
Figure: Data Debtor Koppas Hippatas
In the graph a. group of debtors who have dependents ranges from 0 up to 2 people, more smoothly than the debtor who has dependents ranged from three up to six people. This means that borrowers who have dependent children 0 to 2 are more likely to pay their credit smoothly compared with debtors who have many dependents.
On a graph c. term debtor gets credit ranging from 7 to 24 months. Borrowers may borrow on Koppas Hippatas a term of 7 to 12 months more smoothly than the debtor who borrowed between 14 to 24 months. These indicate the longer term given the greater the risk of loans not paid well by the debtor.
On a graph d. groups of borrowers with credit guarantees in the form of vehicles and machinery, the current higher pay on the credit of not paying his credit. While the debtor granting a deed of sale and purchase of more current in paying his credit. In graph b debtor business is largely operating border and convection in this case because the area is the center of convection and crafting border business.
4.2 Test Variables
The first step in discriminant analysis is to test whether all the independent variables (independent) was significantly different based on the dependent variable (not free), so that can know the variables analyzed feasible and unfeasible.
Test 4.2.1 Average Similarity Group (Equality of Group Means Test)
Equality of Group Means Test is the testing of each independent variable, so they will know whether or not escape these variables for the manufacture of the discriminant model.
Tests of Equality of Group Means
Sig by viewing figures.
- If Sig> 0.05 means there is no difference between groups (no influence).
- If Sig <.05 means that there are differences between the groups (affected)
Of the four variables tested, there are five variables that were significantly different for the two group discriminant, ie, loan, mortgage, and the magnitude Thus, whether or not current customers in paying credit to KOPPAS HIPATAS influenced by loans given to the debtor, the debtor owned mortgage, installment and interest incurred by the debtor.
- Discriminant Model Making
In making the discriminant model, researchers used stepwise discriminant analysis method and the canonical discriminant function. In making this discriminant model included three researchers from the four variables tested on Equality of Group Means Test.
- Inserting and Removing Variables
In this analysis presents the variable anywhere from a variable input that can be entered (entered) in the discriminant equation. In this analysis process was used stepwise (gradual), it will begin with the variables that have a number of F test (statistically) the largest.
Tabel Output Variables Entered/Removed Output Table Variables Entered / Removed
| | Entered Entered | Min. Min. D Squared D Squared | |||||
| Statistic | Between Groups | Exact F Exact F | |||||
| | | Statistic Statistics | df1 DF1 | df2 df2 | Sig. Sig. | ||
| 1 1 | PINJAMAN LOAN | .510 .510 | lancar and tidak lancarcurrent and noncurrent | 17.960 17 960 | 1 1 | 139 139 | .000 .000 |
| 2 2 | TANGGUNGAN Dependent | .903 .903 | lancar and tidak lancar current and noncurrent | 15.786 15 786 | 2 2 | 138.000 138 000 | .000 .000 |
At each step, the variable that maximizes the Mahalanobis distance between the two closest groups is entered. At Each step, the variable That maximizes the Mahalanobis distance Between the two closest groups is entered.
Output Result Tables Eigenvalues
| Eigenvalue Eigenvalue | % of Variance % Of Variance | Cumulative % Cumulative% | Canonical Correlation Canonical Correlation | |
| 1 1 | .229(a) .229 (A) | 100.0 100.0 | 100.0 100.0 | .431 .431 |
a First 1 canonical discriminant functions Were Used in the analysis.
Canonical Correlation measures the closeness of the relationship between the discriminant score of the group (in this case, because there are two types of customers).
Canonical figures of 0.229 and correlation of 0.505. Semakin tinggi harga Eigenvalue , The higher the price an eigenvalue, the better they function in explaining the variables that will be observed.
If the functions in the model is used then the 43.1% variance of the credit variables can be explained by the discriminant model is formed, the remaining balance of 56.9% can be explained by other factors.
Output Result Tables Wilk's Lambda
| Wilks' Lambda Wilks' Lambda | Chi-square Chi-square test | df df | Sig. Sig. | |
| 1 | . | 28.431 | 2 | .000 |
Based on the above table output price obtained Chi-square count of 28 431 with the number of significance 0.000. This indicates that there are significant differences between the two groups in the discriminant models (they are well paying credit obligations on non-current KOPPAS HIPPATAS and in repaying the loan at KOPPAS HIPPATAS.
4:11: The Output Structure Matrix
|
a This variable not Used in the analysis.
Matrix structure above table describes the correlation between independent variables with the discriminant function is formed.
- Two variables that have a fairly tight correlation, in order that is a big variable loans (0.752), and variable mortgage lender owned (0.550).
- From these results, the variable loan is a variable that has the highest coefficient adan is one factor that most distinguishes the behavior of creditors in paying their loans.
- Installment variables and operations are not included in the discriminant analysis model (there is a sign near the point a variable is).
Table Output Canonical Discriminant Function Coefficients
Unstandardized coefficients
The table above is the continuation of the Variables Entered / Removed forming similar functions with multiple regression equation, which is called the discriminant analysis Discriminant Function.
Usefulness of this function to find a case (in this case is a creditor) entered the current group of debtors, or enter the borrower is not current.
Table Output Results Functions at Group Centroids
Unstandardized canonical discriminant functions
evaluated at group means
Therefore there are two types of debtors, then called the Two-Group Diskriminant, where one group has the Centroid (Group Means) is negative and that one group has the Centroid (Group Means) positive. Figures in the table shows the amount of Z separates these two groups.
Output Table of probabilities for Groups Prior Results
| Prior | Cases Used in Analysis | ||
| | | Unweighted | Weighted |
| Smooth | .500 | 68 | 68.000 |
| noncurrent | .500 | 73 | 73.000 |
| Total | 1.000 | 141 | 141.000 |
The above table shows the composition of the 141 respondents, the discriminant model produced 68 current creditors in the group, while the remaining 73 are in a group jam.
Cut Off Score
The results will then be compared with the score cut-off score, is used to determine if consumers go into a group worthy or unworthy.
From the table prior probabilities Groups For consumers who found that the number of feasible and unfeasible respectively 68 and 73 debtors. Therefore, it is associated with a number of group centroids:
Calculation of Z CU (critical number):
Use numbers Zcu (Diskriminating Z score):
- If the discriminant value Zcu each case above, the model can be predicted correctly.
- If the discriminant value of each case under Zcu, then the model can not be predicted correctly or misclasified.
- The calculation process of the model's predictions Created
Results Classification Table Results Output Results
a 68.1% of original grouped cases Correctly classified.
b 67.4% of cross-validated grouped cases classified Correctly.
Table Classification results show that there is a change in the classification of members of groups that occurred between the initial data (Original) with data after the predicted Predicted Group Membership. Teryata there are 17 respondents who stray into non-current group, the group should go smoothly.
This figure is calculated by comparing the number of members of the group that is classified precisely by comparing the number of members of the group who entered with the appropriate classification of the total membership groups were observed. Berikut perhitungan angka ketepatan prediksinya. Following the prediction accuracy rate calculation.
The number of members of the group went right classification:
- Current Credit: 51
- Noncurrent loans: 45
Prediction Accuracy Score = 51 + 45 = 68.08%
If seen from the results of validation (cross-validated) on the code c (under the table) then the prediction accuracy appeared rate of 72.3% together with the results of the above calculation (Original). It can be concluded that the model is feasible to use the discriminant to classify the groups were observed (above 50% means quite feasible to use).
- Comparison of Related Research
The factors that determine whether or not giving proper credit to customers of PT Federal International Finance Branch Bogor.
Variables defined in the selection of credit to customers of PT Federal International Finance Bogor branch consisting of salary, the amount of debt, age, number of installments and the number of children. Based on the results of discriminant analysis, the most influential variable determining whether or not a prospect worthy of a variable while the variable pay and children who are not influential in penenetuan competent or not a customer is a variable debt, angsuan, and age. Figures for the accuracy of the discriminant model which amounted 60.4%.
There are differences in outcome variables included in the model of discriminant analysis in filing due credit to PT FIF Branch Bogor, a debtor must provide a list of salary structures, while in Koppas Hippatas a debtor is a member of the Cooperative.
CLOSING
Conclusion
Based on the analysis and discussion in the previous chapter, the conclusion can be drawn related to the formulation of the problem at the beginning of the case, namely:
- Of the four factors are analyzed to determine the factors that determine the credit worthiness of KOPPAS HIPPATAS, there are two of the most significant predictor of influence. Two of these predictors is owned and loan burden posed by the debtor.
- Predictive model that determines whether or not giving proper credit in KOPPAS HIPPATAS is significant with the predicted rate of 68.1%. Because of above 50% prediction model can be used to determine whether or not worthy of credit on HIPPATAS KOPPAS.
Suggestion
Based on the above conclusion, the authors suggest limited knowledge of the author, as follows:
- KOPPAS HIPPATAS that will give credit to the members can use this discriminant model as a determinant of whether a creditor as current or noncurrent loans in repayment at the Coop. KOPPAS HIPPATAS as a lender does not want to suffer losses, so that before accepting the Cooperative customers can use this discriminant model.
- For KOPPAS HIPPATAS, to invest their funds in the form of loans to members should consider the factors that influence credit payments as consideration for investment decisions. Faktor yang mempengaruhi yaitu tanggungan dan besarnya pinjaman. Factors affecting the mortgage and the amount of the loan. The amount of loan is the most dominant or significant in differentiating the behavior of borrowers in repaying loans, so we need first priority in making lending decisions.
- For subsequent researchers, should perform better and in-depth research, so this result will be more useful in determining the granting of loans.
REFERENCES
Astiko, Sunardi. Astiko, Sunardi. 1996. Pengantar Manajemen Perkreditan . 1996. Introductory Credit Management. Edisi Pertama. Yogyakarta: ANDI. First Edition. Yogyakarta: ANDI.
Hadiwidjaja, Rivai Wirasamita. Hadiwidjaja, Rivai Wirasamita. 1991. Analisis Kredit. Bandung: Pionir Jaya. 1991. Analysis of Credit. Bandung: Pioneer Jaya.
Hair, Joseph. F,et.al., 1998. Multivariate Data Analysis. Fifth Edition. Hair, Joseph. F, et.al., 1998. Multivariate Data Analysis. Fifth Edition. Prentice-Hall, New Yersey. Prentice-Hall, New Yersey.
Kasmir. 2002. Bank dan Lembaga Keuangan Lainnya. Edisi Keenam. Cashmere. 2002. Banks and Other Financial Institutions. Sixth edition. Jakarta: PT. Jakarta: PT. Raja Grafindo Persada. King Grafindo Persada.
Marsuki. 2006. Pemikiran dan Strategi Memberdayakan Sektor Ekonomi UMKM di Indonesia. Edisi Pertama. Marsuki. 2006. Thought and Strategy of Economic Sectors Empowering MSMEs in Indonesia. First Edition. Jakarta: Mitra Wacana Media. Jakarta: Partners Media Discourse.
Santoso Singgih. Buku Latihan SPSS Statistik Multivariat. Jakarta: PT. Elex Media Komputindo, 2002. Santoso Singgih. SPSS Workbook Multivariate Statistics. Jakarta: PT. Elex Press, 2002.
Santoso Singgih. Mengatasi Berbagai Masalah Statistik dengan SPSS Versi 11.5 . Jakarta: PT. Santoso Singgih. Overcoming Problems of Statistics with SPSS Version 11.5. Jakarta: PT. Elex Media Komputindo, 2003. Elex Press, 2003.
Sutojo Siswanto. Sutojo Siswanto. 1997. Analisa Kredit Bank Umum. Seri Manajemen Bank No.2. 1997. Commercial Bank Credit Analysis. Series No.2 of Bank Management. Jakarta: PT Ikrar Mandiri. Jakarta: PT Pledge Mandiri.
Sutojo Siswanto. Sutojo Siswanto. 2008 . 2008. Menagani Kredit Bermasalah. Edisi Kedua. Troubled Credit Menagani. Second Edition. Jakarta: PT. Jakarta: PT. Damar Mulia Pustaka. Damar Honor Book.
Tim Penelitian dan Pengembangan . 2005. Research and Development Team. 2005. Analisis Multivariat dengan SPSS 12. Edisi Pertama. Multivariate Analysis with SPSS 12. First Edition. Jakarta: Salemba Impotek. Jakarta: Salemba Impotek.
Tidak ada komentar:
Posting Komentar