The low income earners have been excluded from financial services due to their limited ability to access credit as they lack good financial options. Their income is volatile, fluctuating daily and they lack reliable ways to harness the power of their...
more
ZBW - Leibniz-Informationszentrum Wirtschaft, Standort Kiel
Signature:
DS 792
Inter-library loan:
No inter-library loan
The low income earners have been excluded from financial services due to their limited ability to access credit as they lack good financial options. Their income is volatile, fluctuating daily and they lack reliable ways to harness the power of their low incomes. This challenge can be met through use of mobile technology to collect data on the socioeconomic activities of the low income earners at minimum costs. The success of the mobile financial services in Kenya cannot be understated coupled with increase in mobile penetration. A mobile based technology for micro-credit already exists through M-Shwari started in the year 2012 on the M-pesa platform to further increase financial inclusion. This paper pro- poses a decision support system that is mobile based for credit scoring, classification and peer group lending of the low income earners in Kenya. This facility is referred to as Mobile Micro Credit System. We hypothesize that, first, mobile micro-credit lending for low income peer groups is similar to that of the conventional individual lending. Second, credit scores and credit quality levels among low income men and women is the same. Third, a mobile based micro credit can further enhance financial inclusion among the low income earners. A comparison is made between the peer groups and individual customers in terms of their credit scores and credit quality levels. The data for the study was extracted from the financial diaries dataset by Financial Sector Deepening. Customers were clustered in peer groups and as individual households based on gender groupings of men only, women only, or men and women. The credit scoring factors were estimated and resulting data trained the hidden Markov model, the classification technique used. The hidden Markov model emit- ted the credit scores and the credit quality levels of the individual households and the peer groups in which the groups had stronger credit scores and credit quality levels compared to those of the individual households. Peer groups for women only, and those of men and women had superior credit scores when compared to men only peer groups. The optimal peer group size for lending is between four and eight members. The current mobile financial services offer a baseline to implement a mobile micro credit service for the low income earners. This is an incentive for financial services providers to consider providing mobile based micro credit loans to low income earners.
