Home

  Page 16 - FIGI - Use of telecommunications data for digital financial inclusion

In paral- Using telecommunications data to enable posi- lel, credit reports showing actual defaults for the tive credit scoring in Brazil same individuals are used as a ground truth to train The Brazilian positive credit scoring agency Quod the user models. (...) The partnership’s product offerings will include cred- performed traditional credit bureau data: ‘Among it insights to complement Quod’s positive scores, those with credit histories, if credit were extended fraud scores to screen credit applications and to the 50% lowest risk prospects according to the on-line transactions, and propensity indicators to credit bureau the default rate would be 9.7%, where- enable digital marketing initiatives. as it would be only 8.3% based on our scoring using Source: See https:// www cignifi com/ post/ manage . . phone records. (...) Telecommunications data can be used not only to 10 While industry participants remain sceptical of reduce risk through credit scoring and profiling, but the ability of any CDR-based model outperforming a to provide information about assets that are being credit score based on data on past repayment histo- financed or insured.

Score: 1468964 - https://www.itu.int/en/publica...n/files/basic-html/page16.html

  Page 19 - FIGI - Use of telecommunications data for digital financial inclusion

KT Corp. has developed the “K-Tel- with WeBank, but the data is used to develop the co Score” and the Credit Rating Delivery Platform artificial intelligence algorithms used by WeBank in (CRDP) as alternative credit rating products using its business. (...) For example, an MNO operating in sub-Saharan a computed value or output, such as a credit score Africa could purchase the CRDP and use the platform or confirmation as to regular locations. The financial to generate credit scores based on their subscriber institution might aggregate the output with other data.

Score: 1458761.9 - https://www.itu.int/en/publica...n/files/basic-html/page19.html

  Impact Evaluation of Brazil’s School Based Financial Education Program

Summary of research findings on policies for improving MSME access to finance  Direct government interventions, such as government banks and credit guarantees, are difficult to implement since design matters a lot and they may be subject to political capture  Additionality – do these interventions support firms that would otherwise not have received credit?  Market-oriented policies, such as collateral registries or credit bureaus, tend to be more successful Credit information sharing can promote MSME lending 6 40 13 0 5 10 15 20 25 30 35 40 45 No credit bureau No credit bureau or registry % o f co u n tr ie s  Introduction of a credit bureau has been shown to increase SMEs’ use of credit  Stronger effects when coverage and scope of information is greater  Governments can put in place the legal framework and/or help lenders overcome coordination failures  Credit registries don’t have the same effect  Focus on larger loans and supervisory purposes Source: World Bank Doing Business (2018) Can psychometric credit scoring provide valuable credit information on MSMEs?  MSME owner takes a “personality” test as part of the loan application that measures  Potential for business success  Integrity or intention to pay back a loan  The Entrepreneurial Finance Lab (EFL) developed such a test based on work in Africa and then also piloted it in Latin America, in collaboration with the IDB  Arráiz, Bruhn, Ruiz, and Stucchi study the viability of the EFL test in a sample of about 1,900 MSME owners who applied for business loans with a bank in Peru Passing a psychometric test leads to increased loan access for MSMEs in Peru Probability of taking out a new loan within 6 months of the application increased from about 55 to 75% at EFL score cutoff (MSME owners with scores above the cutoff were offered a loan as part of the study) Fraction who took out a loan within 6 months BEFORE loan application Fraction who took out a loan within 6 months AFTER loan application 0 .2 .4 .6 .8 1 T o o k o u t n e w S M E l o a n -20 -10 0 10 20 EFL score Sample average within bin Polynomial fit of order 1 0 .2 .4 .6 .8 1 T o o k o u t n e w S M E l o a n -20 -10 0 10 20 EFL score Sample average within bin Polynomial fit of order 1 “Passed” EFL testDid not “pass” EFL test “Passed” EFL testDid not “pass” EFL test Psychometric testing “works” for applicants WITHOUT a credit history Applicants WITHOUT a credit history who were offered a loan based on the psychometric tool were NOT more likely to default on a loan within the following 24 months as those who received a loan based on conventional screening Fraction with loan in default at time of application Fraction with loan in default 24 months after application “Passed” EFL testDid not “pass” EFL test “Passed” EFL testDid not “pass” EFL test BUT psychometric credit scoring does not replace credit bureau information Applicants WITH a credit history who were offered a loan based on the psychometric tool were more likely to default on a loan within the following 24 months as those who received a loan based on conventional screening.

Score: 1458749.4 - https://sdgs.un.org/sites/defa...atements/27106Miriam_Bruhn.pdf

  Page 209 - ITU-T Focus Group Digital Financial Services – Technology, innovation and competition

 Page 209 - ITU-T Focus Group Digital Financial Services – Technology, innovation and competition           Basic HTML Version Table of Contents View Full Version Page 209 - ITU-T Focus Group Digital Financial Services – Technology, innovation and competition P. 209 ITU-T Focus Group Digital Financial Services Technology, Innovation and Competition and used as a basis for developing an alternate credit score, or affecting current credit bureau scoring data. These privacy concerns have garnered the attention of some regulators. 193 10.2 Competition aspects Entities who may be in a position to accumulate data used to create alternative credit scores may potentially use the data to their own advantage by not providing the complete data sets as required to credit bureaus, and/or selectively providing the data only to preferred parties. (...) The CAK 196 study also looks at current practices around consumer control over their transactional data and how this is sold or accessed by third parties such as the usage of mobile credit data to score and award credit offers without consumer consent. 193 See Government Of Kenya (2016) Gazette Notice No. 678: Proposed Market Inquiry And Sector Study On The Kenya Banking Sector-Phase II By Competition Authority Of Kenya, available at https:// goo. gl/ wbqDX6 194 Daily Nation (2016) Govt Launches Study On Mobile Money Practices, available at https:// goo. gl/ h9OumW.

Score: 1415584.8 - https://www.itu.int/en/publica.../files/basic-html/page209.html

  Page 12 - FIGI - Big data, machine learning, consumer protection and privacy

Lenndo, a The Monetary Authority of Singapore recent- 5 fintech firm supporting credit evaluation with alter- ly published Principles to Promote Fairness, Ethics, native data analysis, has partnered with the global Accountability and Transparency (FEAT) in the Use credit agency FICO to make FICO score services of Artificial Intelligence and Data Analytics in Sin- available in India. (...) These seek to apply FAT- 6 data from a consumer's digital footprint to produce style principles specifically to the context of AI and a credit score for those who do not have sufficient machine learning in the financial sector, adding an traditional data on file (“thin file” borrowers) with ethical dimension. (...) The Smart Campaign recently released 7 8 Africa and beyond, using identity proofing and auto- draft Digital Credit Standards, which include a num- mated mobile app that uses credit-scoring engines ber of standards addressing use of data, profiling and to generate credit scores from analysing a custom- automated decisions in digital financial services, and er’s mobile phone bill, text messages, payment his- which are set out in Annex B (Smart Campaign Dig- tory, bank account history (if the person has a bank ital Credit Standards).

Score: 1415506.2 - https://www.itu.int/en/publica...t/files/basic-html/page12.html

  Page 221 - The Digital Financial Services (DFS) Ecosystem

 Page 221 - The Digital Financial Services (DFS) Ecosystem           Basic HTML Version Table of Contents View Full Version Page 221 - The Digital Financial Services (DFS) Ecosystem P. 221 ITU-T Focus Group Digital Financial Services Ecosystem Appendix 2: ACD case studies Company Ant Financial Services Group (Ant) Overview Alipay’s Ant introduced a credit scoring agency, Sesame Credit, which is used for individuals and small businesses by combining public records and financial institution data with Alipay’s marketplace data of more than 300 million consumers and 37 million small businesses. Scores are developed based on five criteria: • Credit history – payment history and indebtedness, including credit card repayment and util- ity bill payments. • Behaviour and preference – online history, including product categories shopped. • Fulfillment capacity – Includes use of financial products and services as well as Alipay account balances. Personal characteristics – personal information, including home address, length of time of residence, mobile phone numbers, etc. • Interpersonal relationships – reflects a user’s friends and their interactions. Sesame Credit scores are being tested for use beyond lending, including high-speed VIP check-in at Beijing’s Capital International Airport , by hotels to allow customers to

  Page 15 - FIGI - Use of telecommunications data for digital financial inclusion

 Page 15 - FIGI - Use of telecommunications data for digital financial inclusion           Basic HTML Version Table of Contents View Full Version Page 15 - FIGI - Use of telecommunications data for digital financial inclusion P. 15 Figure 2 – Development of credit scores with telco data using Machine Learning and agile methodology in Brazil. (...) Geolocation data about a ditionally used in credit scoring, and substantially user, especially when combined with financial data, reduced ‘credit invisibility.’ (...) Initiating larger numbers Telecommunications data also provides valuable of calls rather than receiving them, and making of social information about a user that is useful for pro- calls of long duration, are used in some data models filing and credit scoring. Family and social networks as supporting a higher credit score. can be derived from CDRs or calling plans that fea- Customer and billing records offer direct financial ture special rates for specific individuals (such as data.

Score: 1396876.4 - https://www.itu.int/en/publica...n/files/basic-html/page15.html

  6/6/2016 Shivani Si

Our credit scores have been created through an aggregation and analysis of our public consumer credit data. (...) They don't have a score because there are no formal public records on them -- no bank accounts, no credit histories and no social security numbers. And because they don't have a score, they don't have access to the credit or financial products that can improve their lives.

Score: 1385322.95 - https://www.cepal.org/sites/de...t_history_yet_ted_talk_ted.pdf

  I n t e r n a t i o

Figure 2 - Development of credit scores with telco data using Machine Learning and agile methodology in Brazil. (...) Using telecommunications data to enable positive credit scoring in Brazil The Brazilian positive credit scoring agency Quod is using telecommunications data of the country’s telecommunications operators in partnership with US fintech company Cignifi. (...) The partnership’s product offerings will include credit insights to complement Quod’s positive scores, fraud scores to screen credit applications and on-line transactions, and propensity indicators to enable digital marketing initiatives.

Score: 1384599.9 - https://www.itu.int/en/ITU-T/e...tal%20Financial%20Services.pdf

  Page 215 - The Digital Financial Services (DFS) Ecosystem

 Page 215 - The Digital Financial Services (DFS) Ecosystem           Basic HTML Version Table of Contents View Full Version Page 215 - The Digital Financial Services (DFS) Ecosystem P. 215 ITU-T Focus Group Digital Financial Services Ecosystem Figure 2 – LendingKart Company LendingKart Overview Founded in 2014 and based in Gujarat, India, LendingKart (lendingkart.com) uses more than 1,500 data points to evaluate companies for credit, including ecommerce data, VAT returns, and social media data to produce a financial health score, a marketplace score, a social reliability score, and a statutory compliance score. The marketplace score is derived from e-commerce marketplace sales data, including Flipkart, Snapdeal, Jabong, Amazon, Dehlivery, Power2SME, and M Swipe. (...) Differentiator • Produces multiple scores based on different data sources Summary: Most programs reviewed use a variety of data to support their credit decisioning, finding that the predictive power and usefulness of the mobile data alone is limited.

Score: 1382024.6 - https://www.itu.int/en/publica.../files/basic-html/page215.html