Tác động của tiếp cận vốn vay đối với đầu tư của doanh nghiệp tại tỉnh Điện Biên - Xây dựng thang đo

THE IMPACT OF ACCESS TO LOANS ON BUSINESS’S INVESTMENT IN DIEN BIEN PROVINCE – BUILDING A SCALE TÁC ĐỘNG CỦA TIẾP CẬN VỐN VAY ĐỐI VỚI ĐẦU TƯ CỦA DOANH NGHIỆP TẠI TỈNH ĐIỆN BIÊN - XÂY DỰNG THANG ĐO Dao Thi Van Anh - Tay Bac University vananhdt@utb.edu.vn Abstract: Investment is one of the important decisions of financial management in an enterprise, so it is really necessary to measure the impact of factors on the investment of the business, especially with qualitative factors. By a combination of quantitative and qualitative methods, the article has built a complete scale with indicators measuring the impact of the group of factors on the ability to access loans on enterprises’ investment based on surveyed data at 107 enterprises in Dien Bien province. Keywords: business, investment, ability, scale, loan. Tóm tắt: Đầu tư là một trong các quyết định quan trọng của quản trị tài chính trong doanh nghiệp nên việc đo lường tác động của các nhân tố tới đầu tư của doanh nghiệp là thực sự cần thiết, đặc biệt là với các nhân tố định tính. Bằng sự kết hợp giữa phương pháp định lượng và định tính, bài viết đã xây dựng bộ thang đo hoàn thiện với các chỉ báo đo lường sự tác động của nhóm nhân tố khả năng tiếp cận vốn vay tới đầu tư của doanh nghiệp dựa trên dữ liệu khảo sát tại 107 doanh nghiệp tại tỉnh Điện Biên. Từ khóa: doanh nghiệp, đầu tư, khả năng, thang đo, vốn vay. 1. Introduction Although at present, many government policies are trying to help enterprises improve ac- cess to loans to encourage enterprises to expand investment, especially in fields and industries that benefit the economy, specific regions and localities with many incentives such as the North- western subregion, including Dien Bien province. However, the situation shows that there are still many enterprises that have the need to invest in expanding the scale of production and en- terprises’s activities, but the scale of capital is still limited, the access to loans is still difficult, enterprises had to limit investment and even had to cut their portfolios. Up to now, the previous researches on investment by enterprises in Vietnam are quite lim- ited in terms of quantity, content and research methods. Especially the studies that clarify the im- pact of factors affecting investment of enterprises are mainly done by qualitative research methods. Moreover, although the research on the ability of enterprises to access loans has been mentioned a lot in international studies, it is still relatively new to enterprises in Vietnam. 853 INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 Meanwhile, although not a direct factor, the ability to access loans plays an important role in your investment decisions. Measuring the impact of this factor on business’s investment will help managers have a clear, multidimensional view in investment decisions, local authorities and ministerial authorities will have the basis to refer, propose and develop policies to support busi- nesses. However, the factor of the ability to access loans affecting business investment is accurately measured, completely depending on the scale and measurement indicators. With that in mind, the article uses quantitative research methods to build a set of scales with indicators to accurately measure the impact of access to loans on enterprises’ investments, research in Dien Bien province. 2. Research overview A qualitative factor related to the characteristics of enterprises (not about the macro envi- ronment) studied by the authors is the ability to access loans or financial barriers. There have been a number of studies using quantitative methods to build scales and research on the impact of this factor. However, the object of study (the dependent variable) is not enterprise investment. Although not the main topic of research, the Wit (1996) survey also pointed out the impact of the loan barrier factor on the real estate investment of pension funds and insurance companies in the Netherlands. Not only Wit (1996), but also many other studies also recognize the impact of risks and barriers to debt although that is not their main research topic. For example, Stiglitz et al. (1981), Kochar’s (1999) empirical survey on credit markets in India. Vladimir and Bartlett (2003) based on a sample survey of small businesses in Slovenia to study the financial barriers of enterprises affecting the growth of enterprises. The main barriers identified in the study include factors related to the institutional environment such as the bureau- cracy of local government, external financial constraints such as the very high cost of capital. Research has shown that the growth of enterprises is negatively related to the size of enterprises and that the growth of enterprises has decreased when there are financial barriers. Bigsten et al. (2003) used a quantitative methodology to assess the credit constraints of manufacturing firms in Africa and confirm some scales of credit barriers affecting business per- formance. However, the study of the authors only clarifies the impact on business performance of these enterprises, not investment activities of enterprises. In 2007, research by Galan et al in Spain demonstrated that input costs affect the choice of investment positions and there is a negative relationship between this factor and the investment of enterprises. . In which, input costs are considered as labor costs, transportation costs, capital costs and raw materials costs. The study results also showed that the lower the input costs, the more SME managers in Spain make FDI investment decisions in Latin America. Previous studies have shown the impact of factors of financial risks, barriers to loans on businesses and business results in general, but research results have not been consistent about the impact of this factor. That also shows that, although studies on investment topics, risks and financial barriers of enterprises have been mentioned by many authors, there are still “gaps” that need to be supplemented and clarified. Moreover, there are no studies to measure the impact of the ability to access loans on enterprises’ investment, especially businesses in Vietnam. 854 INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 On the other hand, the measurement of qualitative factors in financial economics is not really popular, especially the measurement of qualitative factors on the ability or capacity of bu- siness activities to affect the investment of enterprises is quite new. Therefore, an in-depth study of the impact of firm’s access to loans on investment is essential. 3. Database and research method Research is done on the basis of a combination of both quantitative and qualitative research methods. Specifically: Qualitative research methods are done by research techniques at the table to identify the temporary scales and indicators, then conduct in-depth interviews with the goal to complete the preliminary scale. In-depth interview technique was performed with 10 enterprises, the intervie- wees were directors or deputy directors of enterprises. Due to limited time and the interview de- pends on the coordination of the interviewees, the interview process is conducted under two forms of direct communication and by phone. The time of each interview lasted about 30 minutes, by the author actively asked and recorded herself. The objective of the in-depth interview is to collect opinions, modify the scale of wording, avoid the situation where the interviewee understands not enough, misunderstands the question or the question is not suitable to the actual situation of enterprises in Dien Bien province. At the end of the in-depth interview process, the survey questionnaire will be completed and used to distribute the questionnaire on a large scale. Primary data will be collected through surveys, which will then be used for analysis with Cronbach’s Alpha’s reliability assessment technique using SPSS 20 software. Analyze Cronbach’s Alpha to check Cronbach’s Alpha coefficients and remove variables with small correlation coef- ficients with total variables. According to Joseph F. Hair et al. (1998), Cronbach’s alpha coeffi- cient of 0.6 or more is a qualified scale, above 0.8 to nearly 1 represents a very good scale. At the same time, the same indicators for measuring a research concept must also be strongly correlated. If the indicator has a correlation coefficient with the total variable greater than 0.3, it is considered satisfactory (in case of less than 0.3, it must be removed from the scale). The process of exploratory factor analysis - EFA is also done by using SPSS 20 software. According to Kim and Mueller (1978) and Kaiser (1974), the Cronbach Alpha reliability assess- ment was performed before EFA analysis to remove trash variables because garbage variables can create dummy factors in EFA analysis. Tools used to analyze EFA include factor load coef- ficients and extracted variance. 4. Scope of the study By the convenient sampling method, in-depth interview technique was done with 10 en- terprises in Dien Bien Phu city. In which, 10 enterprises belong to different types of enterprises with different forms of ownership and operation in different industries. According to Hair et al. (2006), the minimum sample size used for exploratory factor analy- sis should be 100 and the ratio between the number of observations to the number of factors is 5: 1. meaning that each factor needs at least 5 observations. The survey questionnaire was built with 5 factors, so the minimum sample size is defined as: 9 * 5 = 45 observations. The study started a survey from October to December 2019 with 300 enterprise managers 855 INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 (directors or deputy directors of enterprises) in Dien Bien province by the method of non-prob- ability random sampling. Results of the survey have obtained 107 satisfactory survey questions that are larger than the minimum sample size (45 observations). Thus, the sample size is ap- propriate. 5. Research results 5.1. Research results of the table research method and in-depth interviews The research process at the table helps the author to build up the preliminary survey to serve the in-depth interview process. At the end of the in-depth interview, the survey questionnaire was completed with specific, easy-to-understand questions that were consistent with Vietnamese grammar and reality of enterprises in Dien Bien province. On the basis of the scales of Stiglitz and Weiss (1981), Wit (1996), Kochar (1997), Bigsten et al (2003) on financial barriers of businesses, the author develops a scale of factors that reflect the access to loans of businesses. To ensure compliance with the objectives and subjects of the study, some factors have been removed, some elements have changed in words. Wit (1996) uses the factor “Too high fees and charges of banks” to measure barriers to businesses in the borrowing process, this is also one of the factors that reflect the ability access to loans from businesses, be- cause the reasonable cost of loans will help businesses access more and easier capital. However, for the process of processing and interpreting the results from the later research model to be convenient, the scale needs to be neutral, not biased towards good or bad. In addition to banks, businesses can also mobilize capital with many different methods, but they all have to accept a loan cost, so the author renamed the scale to “The cost of borrowing”. In addition, the study of Wit (1996) and Kochar (1997) used the scales “Lack of access to equity” and “Lack of access to venture capital” to measure. However, the results after consulting with experts showed that these 2 scales were not suitable for the purpose of the study, so it was omitted. Details of the scales are shown in Table 1. Table 1. Scale of enterprises’ access to loans Source: Author’s research 856 INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 Sign Description of the scale Note VV1 Amount of borrowing costs (including interest and other costs related to borrowing from banks and other sources) Wit (1996), Kochar (1997), Stiglitz và Weiss (1981), Big- sten et al (2003). VV2 Convenience of loan procedures VV3 The required level of collateral VV4 Disbursement time VV5 The level of “interest” / priority of the banks to small and medium-sized enterprises VV6 Number of credit institutions in the area Wit (1996) VV7 Size of credit institutions in the area Dalberg (2011) VV8 The creditworthiness of the business is assessed by the banks Kira & He (2012) VV9 Financial statements audited Beck, Mak Simovic (2005) 5.2. The results of analyzing the reliability of the scale The study was done with 107 enterprises in Dien Bien province showed that basically, the measure has Cronbach’s Alpha coefficients of 0.713, the correlation coefficient with the total variable of the scales VV5, VV7 and VV8 is less than 0.3, so they were removed. The results of analyzing the reliability of the scale are presented in Table 2. Table 2. Analysis results of reliability of scale Source: Author’s research 5.3. Result of exploratory factor analysis (EFA) The results of the KMO coefficient and Barlett test are presented in Table 3, showing that the KMO coefficient = 0.815> 0.5 and the critical value of the Barlett test is 0.000 which is sat- isfactory, it also means the indicators in the model is correlated with each other. Thus, the sample size is suitable and large enough to implement factor analysis techniques. Table 3. KMO coefficient and Barlett test results The results presented in Table 4 show that 6 indicators ensure the team converges. The total number of Rotation sums of squared loadings of 53,247%> 50% means that the use of factors representing 5 indicators can explain up to 53.247% of explanations of all indicators. Initial Eigen- values = 3,195 in the first factor and starting less than 1 from the 2nd factor shows that the indi- cators in the model converge to only factor. 857 INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 Sign Scale Mean if Item Deleted Scale Variance if Item Deleted Corrected Item- Total Correlation Cronbach’s Alpha if Item Deleted Cronbach’s Alpha = 0,713 VV1 30,91 13,406 0,598 0,648 VV2 31,05 14,385 0,488 0,672 VV3 31,01 12,821 0,632 0,637 VV4 31,37 13,821 0,445 0,676 VV5 31,23 16,275 0,117 0,733 VV6 31,50 14,290 0,399 0,685 VV7 31,81 15,550 0,200 0,722 VV8 31,50 16,686 0,018 0,757 VV9 30,96 12,791 0,700 0,627 KMO and Bartlett’s Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy. Bartlett’s Test of Sphericity Approx. Chi-Square df 15 Sig. ,000 ,815 234,295 15 ,000 Table 4. Total variance explained Total Variance Explained Extraction Method: Principal Component Analysis. Source: Author’s research Factor group scale of access to loans with 6 indicators after testing Cronbach’s Alpha are used to analyze EFA factors. In the process of analyzing the exploratory factor, the author chooses factor loading factor of 0.5, so the indicators with load factor less than 0.5 will be disqualified, so they are not displayed on the results table. Using Varimax Produce rotation method to rotate the original elements to minimize the number of indicators with large coefficients in the same factor. The Varimax rotating factor matrix shown in Table 5 shows that the load factor of VV6 indicators is less than 0.5 so it is removed, the remaining indicators are above 0.5. Table 5. Factor rotation matrix Rotated Component Matrixa Source: Author’s research Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. a. Rotation converged in 5 iterations. 858 INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 Component Initial Eigenvalues Extraction Sums of Squared Loadings Total % of Vari- ance Cumulative % Total % of Vari- ance Cumulative % 1 3,195 53,247 53,247 3,195 53,247 53,247 2 ,917 15,279 68,526 3 ,792 13,195 81,721 4 ,473 7,883 89,604 5 ,368 6,136 95,740 6 ,256 4,260 100,000 Component 1 2 3 VV3 .829 VV4 .828 VV2 .668 VV1 .572 VV6 VV9 .536 6. Conclusion Thus, from a set of scales of factors reflecting enterprises’ ability to access loans with 9 initial indicators, after analyzing the factor of discovery, EFA has formed a set of scales with 5 indicators to explain 53,247% of enterprises’ investment, and 46,753% investment of enterprises explained by other factors. Details of the indicators are shown in Table 6. Research results are really necessary to serve the process of studying the impact of factors’ ability to access loans on enterprises’s investment by quantitative research method. Table 6. Scale of factors of ability to access loans affecting investment of enterprises in Dien Bien province Source: Author’s research Although the impact of the factors of access to loans on investment of enterprises in Dien Bien province has not been clearly confirmed, however, with the research results from in-depth interviews, the author also said that the biggest difficulties that make it difficult for businesses to access loans are interest rates and related procedures before borrowing. Thuc’s study (2017) has shown that most businesses in the Northwest sub-region (including Dien Bien) borrow from credit institutions such as banks and credit funds (accounting for about 70%). Therefore, in order to create conditions for these businesses to borrow capital for investment, the state and local au- thorities should have programs, support or preferential policies on interest rates and loan proce- dures, helping businesses shorten the time and reduce costs when borrowing. At the same time, supporting credit institutions to minimize loan and disbursement procedures to help businesses easily and quickly access loans. REFERENCES 1. Bigsten A., Collier P., Dercon S., Fafchamps M., Gauthier B., Gunning J. W., Soderbom M., Oduro A., Oostendorp R., Patillo C., Teal F and Zeufack A. (2003), “Credit Constraints in Manufacturing Enterprises in Africa”, Journal of African Economies, 12 (1): 104-125. 2. Đặng Công Thức (2017), “Nghiên cứu các nhân tố ảnh hưởng đến lựa chọn nguồn tài trợ vốn của DNNVV - trường hợp tiểu vùng Tây Bắc, Luận án tiến sĩ, trường Đại học Kinh tế quốc dân. 859 INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020 No. Sign Description of the scale 1 VV1 Amount of borrowing costs (including interest and other costs re- lated to borrowing from banks and other sources) 2 VV2 Convenience of loan procedures 3 VV3 The required level of collateral 4 VV4 Disbursement time 5 VV9 Financial statements audited 3. Joseph F. Hair và cộng sự, 1998, Multivariate Data Analysis, 5th Ed, Upper Saddle River, NJ: Prentice Hall. 4. Kaiser HF (1974), An index of factorial simplicity , Psychometrica, 39, 31-36. 5. Kim J-O & Mueller CW (1978), Factor analysis: Statistical methods and practical is- sues, Beverly Hills CA:Sage. 6. Kochar A. (1997), “An empirical investigation of rationing constraints in rural credit markets in India”, Journal of Development Economics, 53 (2): 339-371. 7. Stiglitz J. and Weiss A. (1981), “Credit rationing in markets with imperfect information”, American Economic Review, 71 (3): 393-410. 8. Vladimir B. and Bartlett W. (2003), “Financial barriers to SME growth in Slovenia”, Economic and business review for central and South - Eastern Europe, 5 (3): 161-181. 9. Wit D. P. M. (1996), “Real estate portfolio management practices of pension funds and insurance companies in the Netherlands: a survey”¸ Journal of Real Estate Research, 11 (2): 131-148. 860 INTERNATIONAL CONFERENCE FOR YOUNG RESEARCHERS IN ECONOMICS & BUSINESS 2020 ICYREB 2020