Subjects and Methods of Qualitative Research

A preliminary quantitative study was conducted to assess the reliability and validity of the designed and adjusted scales to suit the conditions in Vietnam. The thesis used a preliminary questionnaire for 100 samples, selected by random sampling method and conducted in March 2017.

The official research was also conducted by the author using a convenient random sampling method, with 570 bank credit officers directly participating in collecting information to serve the bank's lending decisions for SMEs. The research was conducted in 04 provinces of the Northwest sub-region: Hoa Binh, Son La, Dien Bien, Lai Chau.

The research order of the thesis is as follows:

Table 3.1. Research implementation sequence

Steps

Content of implementation
1. Build a preliminary scale	- Through the process of theoretical overview and research overview, choose a suitable scale for the variables. - Results: There are 10 hard information factors and 05 soft information factors. included in the research model.
	- Validity
	* Conduct discussions and in-depth interviews (semi-structured questionnaire) with 10 CBTD,
	02 bank managers and 08 credit managers directly collect
	Customer information and credit rating (understanding of bank credit procedures)
	goods) to ensure the filtering of key influencing factors => Adjust
	scale to suit Vietnamese practice.
2. Evaluation	- Results: 8 key influencing factor groups were distilled and included in the model.
scale through	Figure, divided into 52 observed variables. Complete the questionnaire survey
interview	1st time. Continue to test survey 20 CBTD to ensure the questionnaire is easy to use.
deep and survey	understand, before conducting a preliminary survey
preliminary	- Preliminary survey of 100 credit officers directly lending to SMEs
	* Testing the reliability of the scale, EFA factor analysis
	* For each variable, it is necessary to ensure that the Cronbach's Alpha index is > 0.3 and less.
	total Cronbach's Alpha value.
	* If not sure, review steps (1) overview of the scale, (2)
	translation, (3) expert discussion, (4) removal of some irrelevant indicators.
	- Results: Complete the second survey for official survey

Maybe you are interested!

Steps

Content of implementation

3. Formal quantitative research

- Select 570 random CBTD samples to distribute questionnaires and collect data.

- Distribute survey forms according to the proportion of SMEs in 04 provinces: 38% Hoa Binh; 26% Son La; 20% Dien Bien; 16% Lai Chau.

- Collect and screen ballots, code variables, and enter data into SPSS.

4. Data analysis

- Use SPSS20 software to analyze descriptive statistics, frequency statistics, and evaluate the reliability of the official scale Cronbach's alpha.

- Correlation analysis tests the relationship between independent variables and dependent variables.

- Binary Logistic regression testing of factors affecting lending decisions.

- Test the model's suitability and predictive ability.

5. Results and solutions

- Confirm acceptance/rejection of initial research hypotheses.

- Answer the research objectives.

- Compare and contrast research results with existing overviews.

- Explain the results in relation to the practice of commercial banks and SMEs in the Northwest sub-region.

- Propose solutions.

Source: Author's research

The scale construction process in this study is based on the process proposed by Churchuil (1979). The scale is built based on an overview of research related to factors affecting the credit decision-making process of commercial banks. After building a draft scale for 15 influencing factors, the author used the method of in-depth interviews with 10 CBTDs, 10 managers and directly participated in collecting credit rating information of SME customers in the Northwest sub-region, then the author adjusted the draft scale and observed variables to suit the context of the Northwest sub-region.

Through the results of qualitative research, the draft scale was adjusted to a preliminary scale and was tested with a preliminary sample of 100 credit officers of the Northwest sub-regional bank. The response rate was 100% of the collected questionnaires to ensure reliability (0 error questionnaires), originating from the author himself who conducted the survey directly, based on the relationships of students of the Master's class in Finance - Banking, currently holding leadership positions in commercial banks in the Northwest region. By software

SPSS20, the preliminary scale is adjusted according to: Cronbach's alpha reliability coefficient >0.3 and EFA exploratory factor analysis. Cronbach's alpha coefficient is used first to eliminate inappropriate variables. Observed variables with correlation coefficients between variables and totals below 0.3 will be eliminated (Nunnally et al, 1994). Then, observed variables with factor loading weights less than 0.4 in EFA will continue to be eliminated (Gerbing et al, 1988) and the total variance extracted (>=50%) will be checked. The remaining variables become the complete scale used for official quantitative research.

The official quantitative research used SPSS20 to test the appropriate scale of the model, test the reliability of the model, test the initial hypotheses and the predictive level of the research model.

3.2. Qualitative research

3.2.1. Objectives of qualitative research

Qualitative research is useful in cases where problems are difficult to express in numerical form, or require in-depth research on each research subject. Or when researchers apply models and measures based on theories and studies conducted in different countries, but are the research models and measures suitable for the research context? (Nguyen Dinh Tho, 2013)

The main objective of the qualitative research in the thesis is to distill the factors affecting the lending decisions of commercial banks to SMEs collected by the author in the research overview and to explore new factors related to the practical context of commercial banks and SMEs in the Northwest sub-region of Vietnam, to collect information to make lending decisions of credit officers. Therefore, the author conducted in-depth interviews with practical experts, to initially test the suitability of the research model and the influencing factors selected to be included in the research model.

Thus, the specific objectives of qualitative research are:

First , identify the types of information needed to collect for the process.

make lending decisions.

Second , credit officers assess which type of information has a more important influence on the decision to lend/reject loans to SMEs.

3.2.2. Subjects and methods of qualitative research

To achieve the above goal, the author conducted semi-structured interviews with a group of 20 people, including 2 Deputy Directors of BIDV Bank in Son La province.

La and Viettinbank Hoa Binh province, 08 people are credit department heads, 10 people are CBTD of Agribank, Lien Viet Post Bank, ABbank, MBbank, BIDV, Vietinbank, Vietcombank .... To ensure the representativeness of random interview sampling, the author evenly selected 4-5 people/province of the Northwest sub-region.

Among them, credit officers were carefully selected to ensure representativeness based on a number of criteria: number of years of working in the field of corporate customer credit, credit officers who have directly collected data and assessed customers' credit ratings.

Due to the nature of staff rotation between departments in commercial banks to minimize ethical risks in banking operations, current credit officers or employees (Customer Relations, Risk Control, etc.) who have experience working in credit risk management departments have valuable opinions in credit analysis and credit rating of individual and corporate customers, so choosing to conduct in-depth interviews with this group of subjects does not change the research results.

To make the in-depth interview results valuable, the author used the existing relationships with bank leaders, CBTD who were students of the Master's class K20 in Finance - Banking 2011-2013 at Tay Bac University, and are currently heads of departments in the bank. Based on the existing relationships, the 20 interviewees were willing to cooperate, meet at home or at a cafe, willing to share content related to the process of collecting, processing, and rating corporate customers' credit, as well as boldly adding related personal opinions.

The tool for qualitative research is in-depth interviews. To collect accurate and complete data, the author designed in-depth interview questions (Appendix) and used these open-ended questions to identify factors that influence the credit decision-making process of commercial banks. Questions in the in-depth interview revolve around the following issues:

- Are the factors affecting credit decisions in the global research overview consistent with the reality at Vietnamese commercial banks?

- In the process of collecting credit rating information of SME customers at commercial banks in the Northwest sub-region, what types of information will credit officers collect?

- Among the information collected about SMEs in the Northwest sub-region, which type of information is given priority in the process of making decisions on lending or refusing to lend?

- General assessment of the current situation of local SMEs? Is it easy to access bank loans? What are the reasons why it is difficult for businesses to borrow from banks?

- Comment on the issue: if a business has a good relationship with a government official or a relationship with a bank, does it affect the probability of getting a bank loan?

The interviews were mostly conducted at home after calling to introduce and agree on the time and location (11/20 subjects were willing to be interviewed at home, between 6pm and 8pm; 6/20 subjects were interviewed in their private offices during office hours; the remaining 3/20 subjects were interviewed at a quiet cafe). During the interviews, some bank leaders were not willing to let the interviewer record for various reasons (they were afraid, they had a regulation on speaking in the business, etc.). Therefore, the content of these interviews was fully recorded. The technique used for the interviews in this stage was in-depth interviews through open-ended questions. Based on the results of each person's answers, the author continued to ask questions to clearly exploit the potential attributes that influence the lending decisions of commercial banks to SMEs. The answers from the practical experts will suggest further research contents, supplement useful information in the assessment of SME ratings, contribute to explaining the quantitative research results and propose appropriate recommendations. The notes from the interviews were recorded and compared to find similarities and differences, and draw conclusions on the list of attributes that affect lending decisions.

3.2.3. Qualitative research results

Based on the overview of the literature on information serving the decision-making process for lending to SMEs, the author has proposed two groups of information: hard information and soft information, including 15 factors obtained from the interviewed subjects.

Through synthesizing the results of in-depth interviews with 10 subjects, the following results were obtained:

Firstly , 100% of credit officers agreed that as long as customers satisfy the basic criteria: having a legitimate loan purpose, not being prohibited or restricted from credit granting according to the law, not exceeding the lending limit according to the industry structure, they are able to access bank loans. At the same time, those 15 guiding factors were developed by the interviewees into 52 necessary information attributes based on the actual operations of the bank, the feelings and experiences of the interviewees in the process of lending to SMEs.

“…According to management experience and direct practice related to the corporate customer credit department, BIDV Bank does not first classify SMEs according to the criteria of Vietnamese enterprise law or the World Bank, but has criteria for evaluating and classifying enterprise scale based on each type of business.

or form of ownership. Second, the 15 factors given are all factors that credit officers need to collect and are the basis for classifying businesses according to the internal customer credit rating regulations. Third, it is difficult to divide into qualitative and directional or financial and non-financial, depending on the assessment point of view of the credit officer who directly collects customer information... there are many relationships in life, so sometimes credit officers do not impose compliance with credit procedures or internal credit ratings... there are many special cases of customers such as businesses with good relationships with the government, which are given priority in terms of interest rates, exemptions from collateral, loans greater than prescribed, or other procedures are reduced to create favorable conditions for businesses"

Nguyen The Dung - Director of BIDV Son La branch

“…There are many special cases! Enterprises only need to have good relationships with the government, provincial party committee or bank board of directors to be supported a lot in other appraisal procedures for quick disbursement, not to mention business efficiency or debt repayment ability.... Speaking of SMEs today, the Government's policy is like that and cannot change much, because following the trend of global recession, lending to SMEs is very risky, so the most important thing is to have collateral such as land before discussing business plans, if there is no collateral and no relationships, then 100% of loans will not be disbursed.”

Nguyen Quoc Hung - Director of Vietinbank Hoa Binh province branch

“…The above criteria are all important, but the most important is the financial situation of the business, because almost 100% of them are small businesses, so the understanding of the business owner or business prospects… are all weak, there is no management experience as well as understanding of market developments… so lending is very risky, only when there is a special relationship or collateral such as land can one borrow money from the bank…”

Ta Lan Huong - Credit Officer BIDV Son La Branch

“…According to him, the current internal lending regulations of Vietinbank and other commercial banks highly value the role of non-financial information, but in reality, businesses or individual customers must have collateral and a good relationship with the bank to be able to receive a loan… banks would rather let money die in one place than lend at risk without collateral…”

Nguyen Van Tho - Head of Credit Department, Vietinbank, Son La Province Branch

“…your bank also uses the above 15 criteria to score customers’ credit, the weight of non-financial criteria is always evaluated higher in the process of synthesizing credit scores… the strict credit process goes through many departments and objective controls, so as long as the loan purpose is legal, then consider that it is not in cases restricted by law, does not exceed the lending limit according to the industry structure, has good financial status and has collateral, then the loan is definitely approved…”

Bui Huy Quyen - Credit Officer of LienVietPostBank, Hoa Binh Province Branch

“…I think the 15 criteria above are not enough compared to the lending regulations issued internally by MBank. However, the criteria above are very important and I emphasize the financial factor of a healthy business, especially the audited financial statements are highly appreciated and make it easier to get loans…”

Nguyen Thi Lan - Credit Officer, MBank, Hoa Binh Province Branch

“…Many startups have very good and unique ideas, feasible and effective business plans implemented by capable young people. However, they do not have good social relationships, and do not have collateral, so banks do not dare to invest capital…”

Nguyen Van Lam - Credit Officer of Agribank, Dien Bien Province Branch

“…although ABBank’s credit granting process is very strict and rigorous, the decision to grant a loan or not depends largely on the credit appraisal officer himself. There will be advice and guidance for customers on how to access bank loans. So it is difficult to say that businesses have difficulty borrowing, but it is also easy to say that they have easy borrowing…”

Nguyen Thi Thu Hien - Credit Officer, ABBank, Dien Bien Province Branch

“…when considering whether to lend to a customer or not, the first thing to consider is the legal status and legal purpose of the loan, followed by a feasible business plan, financial situation and finally the collateral. If the business has a feasible business plan, the bank is willing to lend on credit, without collateral, but in return the bank will increase the loan interest rate…”

Nguyen Van Dong - Credit Officer of Agribank, Lai Chau Province Branch

“…in the northwest, all businesses are small, so you can’t expect management experience or audited financial reports, nor do you expect collateral or business forecasts. The most important factor in deciding whether or not to get a loan is whether the business has a good relationship with the government. If the relationship is good, no other factors are needed. If the relationship is not good, then the business certainly does not meet the credit rating criteria, so there is no need to disburse the loan…”

Nguyen Quang Huy - Credit Officer of Vietinbank, Lai Chau branch

Second , the qualitative research results of 10 managers and credit officers of commercial banks in the Northwest sub-region showed that 100% of respondents highly appreciated the role of hard information in collecting customer credit rating information, 40% of respondents mentioned the role of soft information and revealed the relationship network that helps SMEs access bank loans more easily .

The general conclusion after the in-depth interview process shows that the thesis research model is basically suitable for the context and practice of SMEs, there are not many differences in the factors affecting the lending decisions of SME customers of commercial banks in the Northwest sub-region.

After synthesizing and adjusting the model and the first preliminary survey, the author continued to send the survey directly to 20 credit officers at commercial banks in the area to collect feedback from credit officers on the presentation, information quality, and ease of understanding of the questionnaire. Then the author adjusted and completed the scale and the second preliminary survey. This is the practical basis for the author to complete the official survey to conduct the investigation on the total sample.

Thus, after consulting with practical experts on the feasibility assessment of the factors proposed by the author to be included in the initial research model: studying the factors affecting the lending decisions of commercial banks to SME customers in the Northwest sub-region of Vietnam. At the end of the qualitative research process, the author distilled 8 factors that affect the lending decisions of banks: business information factor; financial information factor; collateral information factor; credit history factor; business owner capacity factor; business owner personality factor; social network participation factor; information factor on lending relationship with banks.