Determinants affecting digital financial consumer protection: Evidence from 135 countries in the world from 2014 to 2018

Nội dung text: Determinants affecting digital financial consumer protection: Evidence from 135 countries in the world from 2014 to 2018

1. DETERMINANTS AFFECTING DIGITAL FINANCIAL CONSUMER PROTECTION: EVIDENCE FROM 135 COUNTRIES IN THE WORLD FROM 2014 TO 2018 Dinh Thi Thanh Van, PhD.1 - University of Economics and Business – Vietnam National University, Dao Le Van2 - Fulbright University, Pham Hien Dung3 - University of Economics and Business – Vietnam National University Abstract Asymmetric information in financial digital markets is increasingly becoming a serious problem in the digital era. Consumers of digital finance suffer from asymmetric information compared to financial agents due to the complexity of services and their passive position in collecting, analyzing and processing information. The study offers measures to improve the practice of digital financial consumer protection (DFCP) through quantitative analysis using a sample of 135 countries for the period 2014-2018. DFCP is measured through 8 dimensions, including Access, Product safety and liability, Economic interests, Privacy and data security, Information and transparency, Education and awareness, Dispute resolution and redress, Governance and Participation provided the picture, overview of DFCP during the period 2014-2018. Four groups of factors have positive effects on increasing the protection of financial consumers, including education, institutional innovation, market size (openness), technology infrastructure. Interestingly, the result shows that innovation and economic openness factor over the world in the digital age is a huge opportunity, not a challenge, for increasing well-being of financial consumers. Thereby, governments need to adjust policies that focus on absorbing new technology, encourage innovation and the opening of the economy instead of controlling actions in order to protect their citizen. The study also shows evidence of gender discrimination in digital financial services. Keywords: Digital Financial, Consumer Protection, Technology, Openness, Asymetric imformation JEL code: C81, D82, D18, D53, G14, Q55 1 Corresponding author, gmail: dinhthanhvan@gmail.com 2 Gmail: levandao96kt@gmail.com 3 Gmail: hiendungpham96@gmail.com 1
2. 1. Introduction: Asymmetric information is one of the market failures that requires certain interventions (regulations, laws, education programs, etc). Under the circumstances, several asymmetric- information-based debates occurring in (digital) financial markets were noticeably concerned (The Lifeblood of The Economy). Beginning in the 1970s, Rothschild & Stiglitz (1976) had indicated many drawbacks of supply chain organizations compared to their customers in risk appraisal. In financial markets, financial agents possessed less information about their customers’ ability to repay, loan purpose and ability to manage loans in contrast to their customers. As a result, legal and policy instruments were essential in order to protect financial institutions from steered consumers such as credit score (or bond rating) caps, and maximum loan-to-value (LTV) and debt-to-income (DTI) ratios (Meza & Webb, 1987; Stiglitz & Weiss, 1981; Waller & Lewarne, 1994) as the underwriting criteria. On the contrary, recent researches have exposed some concerns related to asymmetric information, which are closer to financial consumer rights, especially digital financial consumer rights. While the consumers are facing increasingly terminology documents (Burke & Fry, 2019) and difficulties in filtering appropriate information sources, firms welcome data as inputs in their business practices and have strong incentives to collect, use, store, or trade consumer data (Jin & Wagman, 2020). Corresponding with the development of the Big Data system, the banking process of reviewing customer records and monitoring customer behaviour is increasingly tight and simplified (Pộrez-Martớn, Pộrez-Torregrosa, & Vaca, 2018). The weakness in communication between customers and financial agents is also demonstrated by (Gathergood, 2012; Lusardi & Mitchell, 2007; Lusardi & Tufano, 2015). For example, a financial consumer with poor pension planning (70% of responders do not understand how to create a personal financial goal) (NFEC, 2019), opting for unreasonable loans with high-interest rates and borrowing the amount of money above their ability to repay (Bunnell, Osei-Bryson, & Y.Yoon, 2021). The study by Gummi, Finke, Pizatella-Haswell, & Takagi (2019) identifies 6 key weaknesses of digital financial consumers, including: (1) Credit traps and over-indebtedness; (2) Unnecessary burden of credit that fails to meet consumer needs, due to misuse or poor usage of credit products; (3) Misinformed consumers due to lack of transparency; (4) Lack of timely access to required funds; (5) Consumer security and privacy breaches; (6) Fraud liability. Moreover, the asymmetry is aggravated by the rapid development of science and technology which is leading to sophisticated and complexed financial services (Lumpkin, 2010) and the dominance of informal financial sources make information more unpredictable about its authenticity (Seibel, 2001). Economic openness also leads to risks of information transparency and nation/personal security (dynamic hacking) due to the hybrid link between different types of financial services as well as among countries. The complexity of digital financial services is an important reason to reduce consumer trust, according to a survey by James & Jenny (2016), only 5% of respondents choose online loan because financial consumers feel more secure with traditional lending. Black, Lockett, Ennew, Winklhofer, & Mckechnie (2002) and Davison, Watkins, & Wright (1989) also theoretically agree that the complexity of financial products shows inverse proportions 2
3. to consumer choices due to opportunity cost of time to learn products. According to World Bank (2014), only 11% of adults worldwide have legal financial loans (this statistic data drops to 9% for low-income countries, 8% for middle-income countries and 10% for upper-middle-income countries), and 27% for companies worldwide have limited access to finance. The lack of digital financial consumer protection has serious consequences. First of all, there occurs a competitive threat of some new businesses, where consumers obtain a service free of charge and pay by their personal data in return, following with network effects and economies of scale leading to phenomenon “winner-takes-all” that impedes competition (OECD, 2016a; World Bank Group, 2015). Secondly, the term "fragile digital consumers" described in (Colangelo & Maggiolino, 2019) is becoming more and more common with the increasing severity of intra-urban 'financial ecology' (Cuesta-Gonzỏlez, Paredes-Gazquez, Ruza, & Fernỏndez-Olit, 2021). OECD (2017a) believes that vulnerable groups (the elderly, the poor-low-income, low-wealth, less creditworthy households) are at risk of being excluded from digital financial services. Thirdly, the lack of consumer protection leads to a decline in social trust, meanwhile, trust is an important factor influencing transaction costs and social costs Williamson (1979); OECD (2011a). Moreover, a market with a rise in expanding and competing contributes to restricting asymmetric information (Yaseen et al. 2020; Rothbard, 1977; Paul & Cox, 2009). Recent research by Jin & Wagman (2020) also emphasizes the relationship between the increasing information asymmetry and market power then turn the competitive market into an oligopoly. As a result, the relation between asymmetric information, DFCP and market competitiveness deserve careful consideration. Although the documents promoting the protection of financial consumers (regulations, laws, practice) and qualitative research are increasing, quantitative ones are relatively limited. Many efforts have been conducted from developing universal regulations for the FCP (OECD, 2011b) to implementation guidelines (OECD, 2017a, 2018a, 2018b, 2018c) and (World Bank, 2017, 2018), exposing a trend in digital financial consumer protection in the digital era. Efforts to evaluate the situation of financial consumer protection in the digital age can be mentioned (Chen, 2018). Currently, with the best of our knowledge, there has been qualitative research that has successfully measured the digital financial consumer protection index in an international context. The most recent attempt to quantify digital financial consumer protection mentions (Federation of German Consumer Organization, 2017). Therefore, in order to close this research gap, the study has the following major contributions: (1) Determining DFCP index at the level of 135 countries around the world in the period 2014-2018 based on secondary data. Subsequently, we provide an overview of DFCP index in the world. “Digital financial consumers” are individuals using financial service delivered through mobile phones, personal computers, the internet or cards linked to a reliable digital payment system (Durai & G, 2019); (2) Evaluating factors affecting DFCP and provide empirical evidence in the period 2014-2018 in 135 countries, especially, institutional innovation and economic openness. In other words, we focus on answering the question: "Would financial consumers get benefit when the economy is more open and the digital innovation technology becomes more sophisticated?"; (3) 3
4. Contributing a number of policy implications with an aim of DFCP Index improvement all over the world, especially in developing countries. The study is presented as follows. Section 2 provides efforts to measure the current DFCP indicators and identify aspects influencing on DFCP. Section 3 describes the data used in the research model. Part 4 presents the results of the study. Section 5 presents the conclusion and suggestions for further policies. 2. Literature review: Efforts to measure DFCP inclusion are commonly known as researches regarding interviewing experts in a certain extent (e.g. Chen, 2018) examine the overall structure of FCP policies; the financial literacy score developed by OECD (2016b); World Bank (2012) mention consumer complaint and resolution mechanism factors, etc. Federation of German Consumer Organizations (2017) has made efforts to measure DFCF inclusion based on 8 dimensions including 25 criteria with appropriate data proposals to represent those criteria and aspects. Based on the studies of the research group combined with the analytical framework of the Federation of German Consumer Organization (2017), the research measures DFCP based on 8 dimensions with the following representative data (Table 2): Figure 1. 8 Dimensions of DFCP Index and Its criteria Source: (Federation of German Consumer Organisations, 2017) & Author’s collection Based on “Full Model of Financial Capability” (Bunnell et al., 2021), (1) financial skill lead to (2) financial behavior lead to (3) financial situation leads to (4) financial well-being, financial consumers' behaviors are influenced by their knowledge and social concerns (Education & society) 4
5. combined with environmental factors: technology infrastructure, institutional innovation, market size (flexibility, diversity in exchange), etc. Hypothesis 1: Education & social concerns effect to DFCP In the context of increasingly sophisticated commodities, education is an important factor influencing financial consumer protection (Lumpkin, 2010; Robert, 2013). OECD (2017b) emphasize the need for financial consumer protection and digital financial education framework towards the goal of effectively expanding and exploiting financial services in the future. Education, accordingly, restricts potential digital risks concurrently with upgrading the benefits of utilizing financial services. Polat & Abdulsalam (2014) also conclude that consumer enhancement of protection can be increased by three factors which are training, media and electronic means of communications. Research has been conducted within Saudi Arabia banks with data collected from 265 consumers, therefore, the result can reflect consumer perspectives. Advanced DFCP can be relatively simple and convenient such as providing a basic guideline of usage and potential risk information for customers (Example: Google owns YouTube, Microsoft owns Skype, WhatsApp and Instagram belong to Facebook) or providing financial options for retirement planning. With sophisticated and complex financial issues, education helps form and reinforce skills of synthesis and analysis data, improving protective attitudes as well (eg: consumers' willingness to disclose data). In the previous research papers, the evidence of educational improvement to DFCP is relatively ample, such as to digital asset management (Litterscheidt & Streich, 2020); retirement planning; savings accounts; participation in financial markets; less prone to over-indebtedness; internet banking behavior (Andreou & Anyfantaki, 2020). Moreover, education also helps to prevent various potential risks because of enhancing corporate social responsibility (CRS) of firms or reducing costs to detect phishing, hacking attacks and unauthorised use of data. Rửsner, Haucap, & Heimeshoff (2020) utilize data from an overall sample of 179,724 respondents representing the 28 member states of the European Union between 2006 and 2014, concluded an increase of consumer trust by 11% after the implementation of the Unfair Commercial Practice Directive (UCPD). It regulates unfair business practices in the European Union, as part of European consumer law, based on the principle of minimum harmonization. The Directive provides fundamental knowledge so that consumers can defend themselves. Meanwhile, Xu (2019) indicates a positive correlation between social trust and financial inclusion: When trust increases by one standard deviation, financial inclusion index increases by almost 0.5. Hypothesis 2: Technology infrastructure effect to DFCP Digital financial consumer protection (DFCP) index is also influenced by technology infrastructure, for instance, Internet speed, spectrum Internet, Firm-level technology absorption, etc. World Bank (2018) show that Big Data can measure consumer needs, adjust and meet the diverse needs of financial customers in the digital era. Data conversion is also an important factor to protect consumer information. Other supporting researches confirmed the role of Information & 5
6. Communication Technologies (ICT) in dimensions of DFCP Index. Mishra & Bisht (2013) survey 50 poor citizens living in big cities in India point out that the development of technology equipment affects the majority of the population assessing in banking services. Mushtaq & Bruneau (2019) prove that the application of ICT in financial institutions would lead to the development of digital banks. Utilizing the secondary data collected from 54 countries, Sarma & Pais (2011) conclude that digital connection between consumers and banks had a positive effect on financial inclusion. In the regression result by Shamim (2007), the number of internet users significantly enhances financial depth. Hypothesis 3: Institutional innovation improves DFCP The impact of institutional innovation on DFCP has received two popular perspectives, while, some studies suggest that the increasing market power of financial agents through inventing new tools has been creating extremely monopolistic power (eg information processing technology with bigdata has easily discriminated consumers) (Jin & Wagman, 2020), others believe that institutional innovation is a significant factor for financial consumer protection in the digital age. Our argument leans towards the second line of view. First of all, innovation promotes the process of improving the quality and characterization of the products, thus it restricts misappropriation, dynamics hacking and theft (Liu, 2015) as well as reducing transaction costs, synchronizing data, making work-based better for operators and consumers, maximizing the purchase decision, enhancing the customer experience, and saving their time. Marcia & Greta (2014) argue that the development of technology on online platforms enables consumers to increase transactional traffic. According to CGAP (2010), the new platform also enables to systematize financial consumer protection Index. This allows further research into contributing factors and solutions for financial consumer protection practices. World Bank (2012) exemplify effective enforcement of consumer protection, particularly in redress and resolution mechanisms. Pasiouras (2018) quantize the impact of financial consumer protection policies on financial intermediation costs and the influence gap in developing and developed nations is relatively significant. Mariani & Wamba (2020) convince that new big data-based technologies have contributed to creating convenience in digital financial transactions. Another interesting finding by Krafft et al. (2020) shows that with the rapid growth in customer demand (rasing environment and social concern), new business models will be established with a stronger link to their society such as firm-customer data exchange mechanism. An indirect consequence of enhancing innovation through economic growth, service diversification, quality enhancement, reducing transaction costs, available accessibility, etc. is to improve the quality of life of the underprivileged (Aghion & Howitt, 1992; Romer, 1990; Pakes & Griliches, 1980; Oya, Joyce, & Nataliya, 2011; Deaton, 2013) Hypothesis 4: Market size (economic openness) improves DFCP Market size is a noticeable sign reflecting the level of competitiveness. Meanwhile, financial innovation is greater in more competitive financial systems (Thakor, 2012). According to OECD (2009), competitiveness stimulates quality in financial services by reducing the anti-trust of banks know as “too big to fail” and asymmetric information issue (Jin & Wagman, 2020). Further 6
7. development from this perspective, OECD (2011) recommend that financial market should restrict the influence or support of the government, which reduces competitiveness and weakens the market. Yaseen et al. (2020) shows that the market liberalization reforms (in China) had a notable impact on the dynamics of the information environment facing investors in those capital markets. The market power theory address its own failure (including asymmetric information) and the mechanism to promote a sustainable growth have been discussed (see also Paul & Cox, 2009; Rothbard, 1977). Accordingly, the competitive pressure of the market requires firms (digital financial services) to build mutual trust with their customers through giving full and accurate information – consumer- firm data exchange (Krafft et al., 2020). This process can be sluggish but effective (Alkire & Ritchie, 2007). Consequently, market expansion not only has a direct impact on eight dimensions of DFCP index through competition but also is the main factor reducing asymmetric information. In that way, digital financial consumers can also enhance their self-protection capability. Hypothesis 5: Gender discrimination in DFCP Financial consumer protection should concern gender discrimination matter. Utilizing statistical data from the Consumer Federation of America (CFA) and Principal Component Analysis (PCA) model, Fishbein & Woodall (2006) indicate that women are much more likely to receive higher interest rates on subprime loans than men. A conflicting viewpoint by GAO (2018) review found limited evidence of gender price differences, except certain subprime loans. However, these comparisons do not take into account the effect on the differences in product brand, packaging, and other characteristics, which limits the generalizability of the results. Bucher-Koenen, Lusardi, Alessie, & van Rooij (2016) analyze survey responses from German, American, Dutch, indicating that the lack of financial knowledge is more common among women, particularly among young people when the rate of correctly answering all questions between men and women is 38% - 22%. Similar to this theoretical framework, Hsu (2015) runs a regression model from the results of the American survey, showing that financial decision-makers in the family are mainly men. 3. Research method 3.1. Data Step 1: DFCP Index To measure the DFCP Index, we base it on 8 dimensions [table 1]. The indicators below have been used experimentally by different sets of indicators, for example: Global Findex, Network Readiness Index (NRI), ITU Global Cybersecurity, GSMA Mobile Connectivity Index, etc. Data are used as panel data across 135 countries in the period 2014-2018. Details of the data and its source, variables used are described in the table below. Table 2. 8 Dimensions of DFCP and its Proxy Variable Describe Source 7
8. Percentage of respondents, ages 15-60+, have Global Findex database an account at a bank/others financial type ( Access ank.org/financialinclusio n/) 1. The Network Readiness Index (NRI) ndex.org/#highlight Economic 2. Getting credit: Which represents the World Bank, Doing interests highest performance observed on the getting Business credit indicator across all economies ( included in (0-100). ss.org/). It is a trusted reference that measures the ITU Global Product safety commitment of countries to cybersecurity at a Cybersecurity Index and liability global level Dummy variable – Personal Data Protection Policy takes place es/DTL/STI_and_ICTs/I Privacy and note: PCA cannot aggregate Index using CT4D- data security dummy variable, so we do not apply in our Legislation/eCom-Data- actual calculation. However, this is an Protection-Laws.aspx important aspect to consider in further studies. Disclosure index (0-5). The sum of a variety of World Bank, Global Information existing disclosure requirements. We follow Survey on Consumer and Global Survey on Consumer Protection and Protection and Financial transparency Financial Literacy (WB). Literacy. Adult literacy (25%); School life expectancy GSMA Mobile Education and (25%); Mean years of schooling (25%); Connectivity Index awareness Tertiary enrollment (25%) Dispute resolution index (0-0.5-1). We follow World Bank, Global Global Survey on Consumer Protection and Survey on Consumer Dispute Financial Literacy (WB). Protection and Financial resolution and (= 1) if both resolution mechanisms are Literacy. redress available, (= 0.5) if one of the mechanisms is available, (=0) if neither of the mechanisms is available. 8
9. Generic Top-Level Domains (gTLDs) and GSMA Mobile Country Code Top-Level Domains (ccTLD) Connectivity Index per person (20%); Online Service Index score Participation for E-Government (20%); Mobile social media penetration (30%); Mobile apps developed per person (30%) Step 2: Evaluate the impact of factors on the DFCP index. We consider four important factors that influence the DFCP index, including: (1) Education and Social Concern for the development of society (Education & Society); (2) Infrastructure - the availability of technology in the digital era (Technology Infrastructure); (3) Expansion of the market - competitive pressure of the economy (Market size); (4) Institution towards innovation (Institutional Innovation). These data are collected from the Global Competitive Index. In terms of (1) Education & Society, the indicators include the following aspects: Level of primary education, Number of education enrollment (primary level), Government efficiency, Business impact of malaria/Tuberculosis/HIV-AIDS. (2) Technology Infrastructure includes the following aspects: Level of latest technologies, Firm-level technology absorption, FDI and technology transfer, Internet using, Fixed broadband Internet subscriptions. (3) Market size includes the following aspects: GDP, Exports as a percentage of GDP, Domestic market size index, Foreign market size index. (4) Institutional Innovation Index includes the following aspects: capacity for innovation, Quality of scientific research institutions, Company spending on R&D, University- industry collaboration in R&D, Gov’t procurement of advanced tech products, Availability of scientists and engineers. Besides, the research model also controls the variable Gender gap in social media collected from the Mobile Connectivity index. Table 3. Describe Data DFCP Education Infrastructure- Institution- Market Gender index & society technology Innovation size Gap Min -3.4545 2.5826 2.0274 2.1216 1.5495 0.0000 Max 2.6112 6.8212 6.2182 5.7865 6.9359 100.0000 2014 Mean -0.2300 5.4002 3.8782 3.3967 3.8360 70.4491 SD 1.5300 0.9504 1.1413 0.8472 1.1456 30.6450 Min -3.3840 2.7163 2.0670 2.1056 1.7444 0.0000 2015 Max 2.6696 6.8865 6.3649 5.7828 6.9351 100.0000 9
10. Mean -0.1363 5.4690 3.9440 3.4450 3.8998 70.4491 SD 1.5197 0.9554 1.2029 0.8580 1.1049 30.6450 Min -3.2898 2.8611 2.0543 2.2261 1.6688 0.0000 Max 2.7881 6.8683 6.4192 5.7643 6.9778 100.0000 2016 Mean -0.0234 5.4954 4.0500 3.5051 3.9804 70.4491 SD 1.5072 0.8962 1.2219 0.8512 1.1251 30.6450 Min -3.1948 2.8451 1.9348 2.1567 1.6904 0.0000 Max 2.8900 6.8915 6.4133 5.8024 7.0000 100.0000 2017 Mean 0.0917 5.5489 4.1518 3.5474 3.9296 73.0514 SD 1.4834 0.8689 1.2274 0.8476 1.1408 28.2985 Min -2.9698 2.9672 1.9636 2.0782 1.5477 0.0000 Max 2.9395 6.8962 6.4567 5.8212 7.0000 100.0000 2018 Mean 0.2981 5.5967 4.2761 3.5669 3.9964 73.4567 SD 1.4963 0.8605 1.2500 0.8559 1.1227 28.0717 Source: Author’s calculation 3.2. Research model Determining the factors affecting DFCP index is used based on the model: 4 DFCPInt = α + ∑ℎ=1 훽ℎ. 푛ℎ푡 + 훾. 푍푛푡 + 푒푛푡 + 휃푡 Here, DFCPI is deemed as DFCP index measured through Principal Components Analysis (PCA) method. According to OECD (2018d), PCA can summarise a set of individual indicators while preserving the maximum possible proportion of the total variation in the original data set. Largest factor loadings are assigned to the individual indicators that have the largest variation across countries, a desirable property for cross-country comparisons, as individual indicators that are similar across countries are of little interest and cannot possibly explain differences in performance. Sasan et al. (2013) added that PCA’s key advantages are: (i) The alleviation of capacity requirements; (ii) Increased efficiency in a smaller aspects; (iii) Low noise sensitivity. The DFCP index is synthesized through 8 dimensions according to [Table 1]. N is 135 countries with t = 5 (2014-2018) in the sample. Other variables and their coefficients are shown in the Table [2]. 10
11. Table 4. The variables in the model and its expectation Symbol Proxy variable Expectation Education & Educational attainment and social interest. 훽1> 0 Society (X1) Infrastructure – Represents infrastructure related to technology 훽2> 0 Technology readiness. (X2) Market Size (X3) Market expansion and market size reflect the 훽3> 0 competitiveness of the economy. Institution – Institutions towards motivating innovations 훽4> 0 Innovation (X4) Gender gap – Digital financial consumer protection with gender γ Control variable (Z) discrimination taking place [see (Fishbein & Woodall, 2006; GAO, 2018; Hsu, 2015)] e Error term θ Latent variable To solve latent variables, we use the fixed - and random effect model. In particular, the study prioritizes the use of the fixed-effect model due to its advantages. Some robust and sensitive check are proceeded such as Hausman test (see Greene, 2008) described in the appendix. We implement robust standard error via the method of Arellano (1987) for fixed-effect model and the estimator HC3 were suggested by MacKinnon & White (1985) to improve the performance in our samples (135 counties from 2014-2018 as a small sample). The suitability of this model is concluded by Long & Ervin (2000) that HC3 provides the best performance in small samples, as it gives less weight to influential observations. 4. Study results: 4.1. Digital Financial Consumer Protection Index synthesis We use PCA model (principal component analysis) based on 8 dimensions illustrated in table [1]. The detailed results are presented in the following table. 11
12. Table 5. The proportion of variance and cumulative proportion of 8 dimensions PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 Standard 2.2939 2.2939 0.8224 0.8224 0.6022 0.6022 0.3798 0.3067 deviation Proportion 0.6577 0.1069 0.0845 0.0518 0.0453 0.0239 0.0180 0.0118 of Variance Cumulative 0.6577 0.7647 0.8492 0.9010 0.9464 0.9702 0.9883 1.0000 Proportion Source: Author’s calculation Note: The result is done after controlling the scale of each dimension. The analysis results show that 2 determinants PC1 and PC2 contribute to 75% of the total variation and all determinants (except PC2) contribute in the same direction as PC1. We perform the computation of the aggregated DFCP through the multiplication of each component and proportion of variance. The formula is shown below 8 DFCPI = ∑𝑖=1 푃 𝑖 ì 푃 표 표 푡푖표푛 표 푖 푛 푒𝑖 To determine the validity of the DFCP index, we proceed: (1) Examining the DFCP index through several tests (for example, a t-test that assesses the DFCP index of the G20 country group comparing to the developing country group) and comparison with DFCP-related research around the world (eg Chen (2018)); (2) Examining DFCP index in each country on the world map (see figure [2]). The DFCP Index in the G20 group members is expected to be higher than that of the developing countries due to detailed guidelines on practices and policies (see (OECD, 2018a, 2018c)); (3) Examining DFCP index through relevant data sets such as financial Inclusion from World Bank or Global Findex. We expect that recent efforts to develop and improve FCP practices World Bank (2017), DFCP index also witness an increase in the world average. The explanation of total variance according to the two most important determinants is classified into G20 group members (left) and developing nations (right) as below: 12
13. Figure 1. Total variation within G20 group members and developing nations regarding the 2 most important determinants (PC) Source: Author’s calculation The illustration specifies that the DFCP Index of the G20 group members (left) is higher than that of the non-G20 group members and the DFCP Index of the developing country group is worse than the other (right). Moreover, the dispersion of the variation in the G20 group members is also smaller, which is reasonable because the difference in the G20 countries is smaller than that of the rest countries. On the contrary, in the right figure, developing country groups have lower DFCP Index and wider variability than the other region. This also confirms our expectation as developing countries are in the process of improving institution, therefore, the gap among efficiency in DFCP and the development among countries is relatively significant. The T-test is described in detail in Appendix 1. Table 6. DFCP Index of the World, G20 countries, developing countries in the period 2014- 2018 World G20 Developing countries Min Max Mean Sd Min Max Mean Sd Min Max Mean Sd 2014 -3.45 2.61 -0.23 1.53 0.42 2.18 1.37 0.53 -3.11 1.35 -0.88 1.18 2015 -3.38 2.67 -0.14 1.52 0.54 2.2 1.44 0.53 -3.38 1.45 -0.8 1.2 2016 -3.29 2.79 -0.02 1.51 0.62 2.23 1.52 0.5 -3.29 1.61 -0.67 1.19 2017 -3.19 2.89 0.09 1.51 0.73 2.26 1.6 0.48 -3.05 1.77 -0.55 1.16 2018 -2.97 2.94 0.3 1.5 0.81 2.46 1.83 0.46 -2.9 1.88 -0.34 1.16 Source: Author’s calculation Note: The study shows the difference between G20 and developing countries by t-test after controlling the sample size and normal distribution assumptions [Appendix 1]. 13
14. Table [6] shows that: (1) DFCP index in the period 2014-2018 have increased, especially in developing countries, on average, in the period 2014-2018 show an upward movement in growth as 58.82%; (2) While the disparity between the G20 countries tends to decrease, in developing countries and the world in general, it remains unchanged and has a tendency of increasing. The state and movement of DFCP Index in the world in the period 2014-2018 are illustrated on the world map as shown below figure [2]. Figure 2. Digital Financial Consumer Protection Index in the world from 2014 to 2018 Source: Author’s calculation Figure [2] indicates the DFCP index and its changes in the period 2014-2018. Accordingly, dark blue to yellow shows a higher level of DFCP, and a shift in colour blocks represents an increase/decrease in DFCP. Some regions with high DFCP Index are North America, Western Europe, Asia Pacific and Japan. This result is also consistent with research by (Chen, 2018) (Potluri, Sridhar, & Rao, 2020). Meanwhile, many countries in Africa and some ASEAN countries such as Vietnam, Lao PDR, 14
15. Cambodia and Myanmar possess low indicators (12 regions are shown in Appendix 3). This partly reflects the consequence of digital fraud in ASEAN and the immaturity of digital finance (OECD, 2018b). Moreover, according to the report “Financial Capability and Consumer Protection - A way Forward to Financial Inclusion in Africa” shows evidence that a lack of financial literacy on an individual level makes people more vulnerable (Giz, 2010). The colour shift from 2014 to 2018 also demonstrates an improvement in the DFCP index. The most noticeable colour shift as Canada (from orange to yellow); some countries in Sub-Saharan Africa (from dark blue to purple). This confirms many recent reports on an improvement in the DFCP index (European Investment Bank, 2017). 4.2. Model results: The study results are represented in table [7] Table 7. Regression results Dependent variable: Digital Financial Consumer Protection Index (panel 135 countries in the 2014-2018 period) (1) Fixed (2) Fixed (3) Fixed (4) Fixed (5) Random (6) Fixed Education & 0.3143 0.2365 0.1799 Social (0.0613) (0.0399) (0.0458) concern Technology 0.6433 0.5601 0.4677 Infrastructure (0.0394) (0.0348) (0.0379) Institutional 0.6037 0.1291 0.1929 Innovation (0.0667) (0.0495) (0.0560) Market size 1.0755 0.4120 0.7601 (0.0869) (0.0410) (0.0728) Gender Gap 0.0070 0.0055 (0.0012) (0.0015) Intercept -6.0887 (0.2205) Observations 651 651 651 651 651 651 Note: *, , - statistically significant at the 10, 5, and 1%. Table [7] column 1-4 shows that DFCP index is influenced by the following aspects, including Education & Society, Infrastructure-Technology, Institution-Innovation and Market size in the period 2014-2018. In overall, column 5-6 examine these aspects contributing to the DFCP with the additional control gender gap in social media use. According to Fishbein & Woodall (2006) and Mohanty (2014), there is discrimination in the use of digital financial services between men and women, especially in relation to service prices and consumer rights. First of all, education & social interest factors have a positive impact on the DFCP index (column 1,5,6). This positive relationship has abundant evidence. Recently, research by Rửsner et 15
16. al. (2020) evaluated the impact of consumer protection regulations in the digital age on consumer self-protection in financial services has positive results. This study samples on 179,724 people from 28 European countries between 2006 and 2014. Another study of Xu (2019) using a sample of 21,878 observations from 47 countries demonstrates the relationship between social trust, education and financial inclusion [see more (Polat & Abdulsalam, 2014)]. Secondly, Infrastructure - technological readiness has a positive impact on the DFCP index (at column 2,5,6). Much relevant evidence demonstrating the relationship between technology and financial inclusion, accordingly, the study of Mishra & Bisht (2013) observe that mobile technology is an excellent tool to accelerate financial inclusion, particularly in far remote areas. Sarma & Pais (2011) shows a positive relationship between Information and Communication Technologies (ICT) and infrastructure to boost financial inclusion. Shamim (2007) finds a positive relationship between ICT and financial development. More recently, research by Mushtaq & Bruneau (2019) also indicates a positive relationship between financial development and ICT, as a result of that, encourage to reduce poverty and inequality. The study used panel data in the period 2001-2012 within 30 developing countries. Moreover, many significant standpoints related to financial consumer protection in the digital age can come through many different ways, for example, New Forms of Data Processing (World Bank, 2018). Accordingly, new forms of data processing is an essential advantage in the digital era, requiring appropriate technologies to protect financial consumers. Research also points out to many barriers, and technology readiness is one of the major ones. Thirdly, market expansion (openness) contributes to improving the DFCP index shown in column 3,5,6. This research result is not only consistent with the neo-classical theories of competition, but also the results of modern research. Accordingly, enhancing domestic and international competitiveness will contribute to expanding consumer choices in financial services. Competitive pressure in the market will encourage product improvements, product innovations and maintain high service quality (OECD, 2011a; OECD, 2009). For example, in terms of Access (out of 8 dimension), firms (digital financial services) always have a tendency to expand and renovate their financial services for profit purposes couple with enhancing corporate social responsibility (Krafft et al., 2020). Moreover, asymmetric information between customers & financial agents, which is an important cause for the need of DFCP, is also diminished in a more competitive market. The experimental evidence of asymmetric information flow dynamics is indicated by Yaseen et al. (2020). Accordingly, market expansion - enhancing competitive pressure not only improves these dimensions of the DFCP but also prevents asymmetric information, which is considered as an important cause for the need of DFCP. Fourthly, in table [7], column 4-6 shows that a positive relationship between institutional innovation development has a positive effect on the DFCP index. Our research is contrary to speculations that innovations (in terms of technology) handicap well-being of digital financial customers because of increasing the market power to become monopolies firm through the advantaged ability to hold, access and analysis customer’s personal information (Jin & Wagman, 2020). The positive relationship between institutional innovation and DFCP is not only demonstrated 16
17. by direct mechanism (restrict misappropriation, dynamics hacking and theft, reduce transaction costs, synchronize data, make work-based better for operators and consumers, improve the buyer experience, maximize purchasing decisions and save customers' time) but also indirect channel (economic growth, service diversification, quality improvement, and available access for “fragile digital consumers”. Table 8. Research results after controlling the variance. Dependent variable: Digital Financial Consumer Protection Index (panel data with 135 countries in the 2014-2018 period) (7) Fixed - Arellano (8) Fixed – HC3 0.1799 0.1799 Education & society (0.0554) (0.0566) Infrastructure – 0.4677 0.4677 Technology (0.0425) (0.0435) 0.0425 0.1929 Institution - Innovation (0.0735) (0.0746) 0.7601 0.7601 Market size (0.0836) (0.0882) Gender Gap 0.0055 0.0055 (control variable) (0.0038) (0.0049) Observations 651 651 Note: *, , - statistically significant at the 10, 5, and 1%. 5. Conclusion: In the digital age, consumers are facing more and more disadvantages in participating in financial services due to the consequence of asymmetric information, so it is necessary to take measures to improve digital financial consumer protection. Our research has partially addressed this research gap through: Firstly, the DFCP index is measured at an international level with samples from 135 countries for the period 2014-2018. DFCP index is measured on 8 fundamental aspects (table [1]). Besides, the study also proposes an overview of DFCP index in the world map and changes throughout the period (see figure [1]). 17
18. Secondly, the study provides theoretical and empirical evidence on the factors affecting the DFCP index. In particular, the research team especially emphasizes the group of factors on technology readiness and market openness with a significant impact. Financial consumer protection in the digital era is essential not only for social stability, market failure (asymmetric information) but also for long-term sustainable growth. To improve digital financial consumer protection, the government can intervene through several ways such as: Raising consumer awareness through fundamental guidance and encourage self-study; Improving infrastructure related to technology readiness (internet spectrum, ) to build new platforms in digital financial consumer protection [e.g. data transformation (World Bank, 2018)]; Strengthening competitive pressure on the economy to reinforce product quality (expanding economic trade, signing bilateral - multilateral agreements, e.tc); Supporting innovate programs to enhance creativity in the market (for example: Investing in R&D research in universities). Our research contributes to shedding light on financial agents' market power asymmetry compared to their customers. Instead of straining the firm-customer conflict, innovation propitiates it through a various mechanism, for example, firm-customer data exchange channel, accordingly, firms provide information to customers efficiently, in return their customer informs sufficiently and accurately personal data by mutual trust and profitable relationship. Furthermore, innovation contributes to improving product quality, diversifying types of services, reducing transaction costs, propitiating customer decision-making and saving their time, etc. Interestingly, expanding market by enhancing economic competition is an important factor supporting asymmetric information restrictions, and concurrently a major reason leading to the urgency in DFCP. Moreover, there are many evidences that market expansion strongly affects 8 dimensions of DFCP. The lack of DFCP mechanisms has also led to many consequences that reduce market competitiveness, such as "winner take all" or data business (OECD, 2016a; World Bank Group, 2015). Therefore, this study is an important evidence that complements the importance of market expansion and government competitiveness enhancement for a sustainable growth. The research once again has supplemented the evidence for policies promoting open-economic policies. In the field of data sharing, research by Potluri et al. (2020) has shown the disadvantages of data localization regimes (such as China and Russia) on the economy. Besides, enhancing digital finance education combined with technology infrastructure is an important foundation to absorb the advantages of innovation (Artificial intelligence, learning machine) and the inevitable wave of globalization in the digital age, which is the border is just a territorial concept. Finally, we have not yet considered the impact of regional factors, whereby developing countries may be at a disadvantage due to poor education levels and limited technology infrastructure. The paper opens up further research directions: (1) Determining factors affecting DFCP in a specific regional context. For example: "Are developing country financial consumers get benefiting from globalization in the digital age?" and (2) if so, “Which levels will the economic reach expand in order to reach the optimal DFCP?” If no, (3) any conditions (education or technology readiness) 18
19. for integration are the factor that enhances DFCP; Thereby, (4) Studies can evaluate specific policies of countries in improving DFCP. Some data limitations: Due to the limited data related to DFCP, the study confirms the following limitations of data: (1) Representativeness. Of eight dimensions of the DFCP index, there are some indicators that may be more representative (than the data used in the study), however, are not used due to the incomplete statistical data (135 countries from 2014 to 2018). For example, the A4AI data is more representative in terms of the access aspect but it is only available in developing countries. Therefore, we use an equivalent data as % have an account at the bank or another type of financial institution or personally using a mobile money service (in the past 12 months) of the Global Fintex dataset; (2) Missing data problem: Some data are collected every 2 years or 3 years, as a result with some missing data we estimate through growth rate in period or using the most recent data in case of discrete data. These estimates are reasonable because of the absence of major shocks in the database for the 2014-2018 period. Appendix Appendix 1. The distribution of DFCP Index between G20 countries and Developing countries is not extreme far from normal distribution with skew.2SE(G20) = -0.347 kurt.2SE(G20) = -1.382 skew.2SE(Developing countries) = -0.877 kurt.2SE(Developing countries) = -1.377 T-test of DFCP Index between G20 countries and Developing countries (variance are not equal with var test - p-value DFCPDeveloping countries) Appendix 2. Test results of model 6 (fixed) Test Null Hypothesis P-value Testing for fixed effects H0: OLS better than fixed p-value < 2.2e-16 Hausman Test H0: RANDOM better than fixed p-value = 2.951e-12 Time-fixed effects testing H0: No time-fixed effects are p-value = 0.1388 needed Heteroskedasticity testing H0: Homoskedasticity p-value = 2.714e-12 (Breusch-Pagan) Note: Due to heteroskedasticity phenomenon, we perform robust standard error in model (7) and (8) in table [ ]. 19
20. Appendix 3. 12 Divided economic regions References Aghion, P., & Howitt, P. (1992). A Model of Growth Through Creative Destruction. Econometrica, 60(2), 323-351. Alkire, S., & Ritchie, A. (2007). Winning Ideas: Lessons from free-market economics. Retrieved from WORKING PAPER NO. 6.OPHI gratefully acknowledges the support of the International Development Research Council (IDRC) for this research: Andreou, P. C., & Anyfantaki, S. (2020). Financial literacy and its influence on internet banking behavior. European Management Journal. Arellano, M. (1987). Computing Robust Standard Errors for Within-Groups Estimators. Oxford Bulletin of Economics and Statistics, 49(4), 431-434. Black, N., Lockett, A., Ennew, C., Winklhofer, H., & Mckechnie, S. (2002). Modelling consumer choice of distribution channels: An illustration from financial services. International Journal of Bank Marketing, 20, 161-173. Bucher-Koenen, T., Lusardi, A., Alessie, R., & van Rooij, M. (2016). ow financially literate are women? An overview and new insights. NBER Working Paper No. 20793. Issued in December 2014. NBER Program(s):Economics of Aging. Bunnell, L., Osei-Bryson, K.-M., & Y.Yoon, V. (2021). Development of a consumer financial goals ontology for use with FinTech applications for improving financial capability. Expert Systems with Applications, 165, 113843. Burke, M., & Fry, J. (2019). How easy is it to understand consumer finance? Economics Letters, 177, 1-4. CGAP. (2010). Financial Assess 2010. Consultative Group to Assist the Poor. Retrieved from 20
21. Chen, T.-J. (2018). An International Comparison of Financial Consumer Protection Republic of Korea. Colangelo, G., & Maggiolino, M. (2019). From fragile to smart consumers: Shifting paradigm for the digital era. Computer Law & Security Review, 35, 173-181. Cuesta-Gonzỏlez, M. l., Paredes-Gazquez, J., Ruza, C., & Fernỏndez-Olit, B. (2021). The relationship between vulnerable financial consumers and banking institutions. A qualitative study in Spain. Geoforum, 119, 163-176. Davison, H., Watkins, T., & Wright, M. (1989). Developing New Personal Financial Products – Some Evidence of the Role of Market Research. International Journal of Bank Marketing, 7(1), 8- 15. Deaton, A. (2013). The great escape: Health, wealth, and the origins of inequality. Durai, T., & G, S. (2019). Digital Finance And Its Impact on Financial Inclusion. European Commission. (2018). Consumer Markets Scoreboard: making markets work for consumers - 2018 edition. European Investment Bank. (2017). Digital Financial Inclusion in sub-Saharan Africa Retrieved from Federation of German Consumer Organisations. (2017). Indicators of consumer protection and empowerment in the digital world. Markgrafenstr. 66 10969 Berlin, Germany. Fishbein, A. J., & Woodall, P. (2006). Women are Prime Targets for Subprime Lending: Women are Disproportionately Represented in High-Cost Mortgage Market Retrieved from Consumer Federation of America (CFA): Fishbein, A. J., & Woodall, P. (2006). Women are Prime Targets for Subprime Lending: Women are Disproportionately Represented in High-Cost Mortgage Market Retrieved from Consumer Federation of America (CFA): GAO. (2018). Consumer Protection: Gender-Related Price Differences for Goods and Services. Retrieved from Washington, DC 20548: Gathergood, J. (2012). Self-control, financial literacy and consumer overindebtedness. J. Econ. Stud. Psychol., 33, 590-602. giz. (2010). Financial Capability and Consumer Protection – A way Forward to Financial Inclusion in Africa. Retrieved from Accra, Ghana: Greene, W. H. (2008). Econometric Analysis. Stern School of Business. Retrieved from New York: Gummi, M., Finke, L., Pizatella-Haswell, R., & Takagi, Y. (2019). Consumer Protection In Financial Technology Risk Unbanked And Underbanked Consumers Face In Digital Credit And The Technological And Regulatory Approaches To Mitigate Risk. Retrieved from 21
22. Hsu, J. W. (2015). Aging and Strategic Learning: The Impact of Spousal Incentives on Financial Literacy. Journal of Human Resources, 51(4), 1036-1067. James, H., & Jenny, W. (2016). Customer Experiences of Using Mortgage Brokers Market Research Report prepared for the Mortgage & Finance Association of Australia. Retrieved from Mortgage and Finance Association of Australia: Jin, G. Z., & Wagman, L. (2020). Big data at the crossroads of antitrust and consumer protection. Information Economics and Policy, 100865. Krafft, M., V.Kumar, Harmeling, C., Singh, S., Zhu, T., Chen, J., . . . Rosa, E. (2020). Insight is power: Understanding the terms of the consumer-firm data exchange. Journal of Retailing. Litterscheidt, R., & Streich, D. J. (2020). Financial Education and Digital Asset Management: What’s in the Black Box?*. Journal of Behavioral and Experimental Economics, 87, 101573. Liu, Y. (2015). Consumer protection in mobile payments in China: A critical analysis of Alipay’s service agreement. Computer Law & Security Review, 31(5), 679-688. Long, J., & Ervin, L. (2000). Using Heteroscedasticity Consistent Standard Errors in the Linear Regression Model. The American Statistician, 54, 217-224. Lumpkin, S. (2010). Consumer Protection and Financial Innovation: A Few Basic Propositions. OECD Journal: Financial Market Trends, 2010/1. doi: 2010-5km7k9tp2jxv Lusardi, A., & Mitchell, O. S. (2007). Baby boomer retirement security: The roles of planning, financial literacy and housing wealth. J. Monetary Econ, 54, 205-224. Lusardi, A., & Tufano, P. (2015). Debt literacy, financial experiences, and overindebtedness. J. Pension Econ. Financ., 14, 332-368. MacKinnon, J., & White, H. (1985). Some Heteroskedasticity-Consistent Covariance Matrix Estimators with Improved Finite Sample Properties. Journal of Econometrics, 29, 305–325. Marcia, P., & Greta, B. (2014). In the Fast Lane: Innovations in Digital Finance. International Finance Corporation. Retrieved from WorldBank Group: Mariani, M., & Wamba, S. (2020). Exploring how consumer goods companies innovate in the digital age: The role of big data analytics companies. Journal of Business Research, 121, 338-352. Meza, D. d., & Webb, D. C. (1987). Too Much Investment: A Problem of Asymmetric Information. The Quarterly Journal of Economics, 102(2), 281-292. Mishra, V., & Bisht, S. S. (2013). Mobile banking in a developing economy: a customer-centric model for policy formulation, Telecommun. Policy, 37(6), 503–514. Mohanty, V. L. (2014). Gender & Consumer Rights: Gender Discrimination as a consumer. Mushtaq, R., & Bruneau, C. (2019). Microfinance, financial inclusion and ICT: Implications for poverty and inequality. Technology in Society, 59, 101154. 22
23. OECD. (2009). Competition and Financial Markets OECD. (2011a). Competition Issues in the Financial Sector. OECD. (2011b). G20 High-Level Principles On Financial Consumer Protection. Retrieved from OECD. (2016a). Big Data: Bringing Competition Policy to the Digital Era – Background Note by the Secretariat. Retrieved from OECD. (2016b). OECD/INFE International Survey of Adult Financial Literacy Competencies Retrieved from OECD, Paris: OECD. (2017a). Considerations For The Application Of The G20/OECD Highlevel Principles On Financial Consumer Protection To Digital And Alternative Financial Services. Retrieved from DAF/CMF/FCP/RD(2017)2/FINAL: OECD. (2017b). G20/OECD INFE report on digitalisation, consumer protection and financial literacy. Ensuring financial education and consumer protection for all in the digital age. Retrieved from OECD. (2018a). Considerations for the Application of the G20/OECD HighLevel Principles on Financial Consumer Protection to Digital and Alternative Financial Services. Retrieved from OECD. (2018b). Financial Inclusion And Consumer Empowerment In Southeast Asia. Retrieved from OECD. (2018c). G20/OECD Policy Guidance on Financial Consumer Protection Approaches in the Digital Age. OECD. (2018d). Handbook on Constructing Composite Indicators. Oya, A., Joyce, I., & Nataliya, M. (2011). Consumer Protection Laws and Regulations in Deposit and Loan Services: A Cross-Country Analysis with a New Data Set. Retrieved from NFEC. (2019). Top 3 Most Missed Questions, As of January 1, 2019. National Financial Educators Council National Financial Literacy Test Results. Pakes, A., & Griliches, Z. (1980). Patents and R and D at the Firm Level: A First Look. NBER Working Paper No. 561. Paul, A. C., & Cox, S. (2009). Literature and the Economics of Liberty: Spontaneous Order in Culture. Ludwig von Mises Institute. Pộrez-Martớn, A., Pộrez-Torregrosa, A., & Vaca, M. (2018). Big Data techniques to measure credit banking risk in home equity loans. Journal of Business Research, 89, 448-454. Polat, A., & Abdulsalam, A. A. (2014). Consumer Protection in Banking: Investigating the 10 High Level Principles of G20 in Saudi Arabia. Journal of Applied Finance & Banking, 4(3). 23
24. Potluri, S. R., Sridhar, V., & Rao, S. (2020). Effects of data localization on digital trade: An agent-based modeling approach Telecommunications Policy, 44(9), 102022. Robert, C. M. (2013). Observations on Financial Education and Consumer Financial Protection. Life-Cycle Investing: Financial Education and Consumer Protection. Microeconomics: Life Cycle Models & Behavioral Life Cycle Models eJournal, 13-15. Romer, P. (1990). Endogenous Technological Change. Journal of Political Economy, 98(5), S71-102. Rửsner, A., Haucap, J., & Heimeshoff, U. (2020). The impact of consumer protection in the digital age: Evidence from the European Union. International Journal of Industrial Organization, 102585. Rothbard, M. N. (1977). Power and Market: Government and the Economy. Rothschild, M., & Stiglitz, J. (1976). Equilibrium in Competitive Insurance Markets: An Essay on the Economics of Imperfect Information. The Quarterly Journal of Economics, 90(4), 629-649. Sarma, M., & Pais, J. (2011). Financial inclusion and development. J. Int. Dev., 23(5), 613- 628. Sasan, K., Shahidan, M. A., Azizah, A. M., Mazdak, Z., & Alireza, H. (2013). An Overview of Principal Component Analysis. Journal of Signal and Information Processing, 4, 173-175 Seibel, H. D. (2001). Mainstreaming Informal Financial Institutions. Working Paper, No. 2001,6, Universitọt zu Kửln, Arbeitsstelle fỹr Entwicklungslọnderforschung (AEF), Kửln. Shamim, F. (2007). The ict environment, financial sector and economic growth: a cross- country analysis. J. Econ. Stud., 34(4), 352-370. Stiglitz, J. E., & Weiss, A. (1981). Credit Rationing in Markets With Imperfect Information. American Economic Review, 71(3), 393-410 Thakor, A. (2012). Incentives to innovate and financial crises. Journal of Financial Economics, 103(1), 130-148. Yaseen, S. A.-Y., John, T. B., & Arav, O. (2020). Liberalization and asymmetric information flow dynamics in the Chinese equity markets. The Journal of Economic Asymmetries, 21, e00151. Waller, C. J., & Lewarne, S. (1994). An expository model of credit rationing. Journal of Macroeconomics, 16(3), 539-545 Williamson, O. (1979). Transaction-cost economics: The governance of contractual relations. Journal of Law and Economics, 22, 233-261. World Bank. (2012). Good practices for financial consumer protection. Retrieved from World Bank. (2014). Enterprise Surveys. World Bank. (2017). Good Practices for Financial Consumer Protection. 1818 H Street NW, Washington DC 20433. 24
25. World Bank. (2018). Financial Consumer Protection and New Forms of Data Processing Beyond Credit Reporting. Retrieved from he World Bank 1818 H Street NW Washington DC 20433, USA: World Bank Group. (2015). Digital Dividends: World Development Report 2016. Retrieved from 19; BEUC, ‘A Consumer-Driven Digital Single Market’: World Economic Forum. (2016). The Global Information Technology Report 2016. Retrieved from Xu, X. (2019). Trust and Financial Inclusion: A Cross-country Study. Finance Research Letters, 35, 101310. 25