How Blockchain, Big Data and AI are helping developing countries’ businesses to access financing

The question of small and medium business (SMB) support in developing economies has never been so vital as today. Due to the fall of oil prices the majority of developing countries in Africa and Latin America are experiencing a long-lasting recession. SMBs are eager for money, but they don’t have access to financing for business development, because of high interest rates and the absence of relevant data for standard scoring models. This situation presents FinTech startups with huge opportunities, using evaluation tools for borrowers in developing countries by using alternative credit scoring methods based on artificial intelligence, Big Data and Blockchain.

According to Doing Business Report 2018, globally twenty-four economies implemented reforms improving their credit information systems in 2016/17. Among those developing economies are the most active, aiming to improve the credit information sharing within new credit bureaus. However, in a number of countries the financial systems lack unified trusted tools and institutions; the loan market is small and individual’s’ credit history are either insufficient for validation or muddled due to lack of national IDs. For example, in Bolivia, credit bureau covers only 50% of adults, in Ghana 16%, in India 43,5%, in Kenya 30.4%, and in Nigeria only 7.8%.

As a result, traditional financial institutions cannot provide people in developing countries with loans, due to absence of credit history. At the same time, penetration of mobile phones is quite high: for example, while in Africa 80% of the population does not have bank accounts, but 63 out of 100 people use mobile phones. This year has seen a rise of FinTech startups that use technology to open up new opportunities for small and medium business lenders in developing economies to obtain secure financing via their mobile phones.

How Machine learning powers credit scoring and credit risk assessment

To understand the opportunity and potential solutions, one has to take a step back though and understand the fundamentals of the process. Traditionally, scoring is considered on the basis of a number of facts about a person or a business. What is the person’s occupation? Where does he or she live? What is the total salary? Combined and weighted all this data is turned into a solvency score. But new FinTech startups can break this stereotype out leveraging machine learning algorithms and mobile technology. Any piece of data can be used for a person’s evaluation. Computers can select and analyse data from open sources like social networks, process a borrower’s questionnaire, and even check some physical and behavioral data to lower risks and make loans more affordable.

For example, KiaKia has recently presented an alternative credit scoring, which aims to drive down the interest rate for unsecured loans. Leveraging psychometric, big-data, machine learning and digital forensics for its proprietary credit scoring and credit risk assessment algorithm allows securely providing direct and personal and business loans to millions of individuals and SMEs without any valid credit information.

Another interesting case is the partnership of MicroMoney and Karma: MicroMoney will access repayment data from Karma, and Karma will use MicroMoney’s algorithms to establish a borrowers’ creditworthiness. MicroMoney uses machine learning algorithms and big data tools and stores results securely in the blockchain. The company has recently raised $10m via an ITO. On the other side of the partnership sits Karma, a platform that provides credit scores to users but also serves as a portal for people to search and apply for various financial services like loans, credit cards and insurance. It also allows its users to lend to and borrow from each other. This is particularly relevant as peer-to-peer business lending is the third-largest business models in African Fintech. Over a two-year period between 2014 and 2015 it $16 million in volume. While this does not seem much compared to more mature FinTech markets, it is a growing trend and we are likely to see more collaborations of this kind to come.

“We want to help companies using a Blockchain technology to scale their client bases effectively by means of credit histories bureau with information on millions of people who aren’t served by banks now” as Anton Dzyatkovsky says, the CEO of MicroMoney, a global FinTech blockchain company and lending services provider based in Singapore. “We want that these people to become a part of the new global decentralized economy. We provide access to such data through the API. When we are done with the blockchain part of database, the entire pool of existing credit stories will be transferred to a blockchain network. Thus, we will ensure safety and reliability of collected information. Our app uses more than 10 000 parameters to assess credit risk of the individual in 15 sec via neural nets and AI’.

Big Data analysis tools enhance validation of psychometric tests

A vital part of the new evaluation models lies in the proper use of Big Data analytics. For instance, the psychometric practice for scoring started in 2015, when Entrepreneurial Finance Lab came up with a testing application, which assessed potential borrower’s’ ability and willingness to pay loans back. Such organizations as IDB and MIF have been working with the Entrepreneurial Finance Lab (EFL) and arranged a real methodology test in Peru. The results showed, that psychometrics tests were efficient to assess creditworthiness of a person without credit history.

Since that time, psychometrics tests were implemented by many financial organizations for borrowers’ reliability verification.  However, the psychometric approach has several problems. The results can be biased and depend on circumstances.

Therefore, startups are focusing on finding solutions that address these problems in order to open new ways for FinTechs in Africa, Asia, Latin America and other regions. A first step could be to store the credit history and psychometric information in a distributed network, i.e. blockchain and accessed on demand. Blockchain can offer a reliable accessible and affordable way for creating a global “credit bureau” to store a wide spectrum of scoring-related information. One of the startups working on such a solution, for example, is Human Discovery Platform, which aims to collect various psychometrics tests, saving the results on a blockchain and providing companies with the tools to store and analyze people’s behavioral data. Thus, any FinTech startup can use ever-growing psychometric base and implement some AI or ML algorithms to deepen the scoring model. Even if a potential customer does not have an address or a credit history, a number of psychometric tests could provide better insight as to the credibility of a borrower compared to traditional data, which can often be misleading. The results of any such test can then be saved on a blockchain until the customer applies for a loan at any time in the future. “Big Data analysis tools and Blockchain technology opens up opportunities to get a explicit portrait of the borrower, and continuously complement it with personal data”, explains Timur Karimbaev, CEO of Human Discovery Platform. By doing so, FinTech startups that seek to enter this space but also traditional institutions that are open to innovation can access loans and provide better services than ever before.

Conclusion

The rapidly growing FinTech sector continues to increase its foothold in traditional banking markets. Bloomberg reported that FinTech investments in Asia surged to $10.5 billion in the first nine months of 2016 in comparison with $4.3 billion in 2015. This is further underlined by a statement from the Basel Committee on Banking Supervision from August 2017, which stressed that FinTech has the potential to change traditional banking business models, structures and operations. FinTechs are more flexible and agile, and they are ready to use anything: AI, Machine Learning, blockchain or any data source, to validate the borrower and enter new markets securely. The more they use innovative tools, the more mature and precise these new methods will become. And if future looks bleak for incumbents in mature financial markets, their situation in developing countries, looks like they have already lost the fight..

 


This article is a guest post from Julia Zhidkova, a storyteller and responsible for PR at ICO Mill, Moscow-based consulting agency for blockchain companies. Julia is a qualified strategic communications
professional having worked for years with international IT, industrial and state companies at AGT Communication Group, one of the largest communication firms in Russia, part of the leading global network PROI Worldwide. Julia enjoys writing about ICOs, innovative businesses, technology and communication. You can find Julia on LinkedIn. If you want to know more about FinTech, check out our dedicated section here.