M-Shwari loans help Kenyan households weather shocks

Kenyan households just above M-Shwari's credit-score cutoff were 6.3 percentage points less likely to forgo expenses after a negative shock than those just below it.

  • 2 a person

    Households just eligible for loans were 6.3 percentage points less likely to forgo expenses after a shock, a measured gain in resilience near the cutoff.

Depth runs from 1, a convenience or time saved, to 5, a life or a livelihood preserved.

Status
Reported benefit.
Evidence

foundstudiedreportedclaimed

An independent peer-reviewed study supports the benefit.

Compared with

Households with credit scores just below the M-Shwari cutoff. Measured.

Without the system
approximately 68% of the control group reports having to forego some expenses
With it
6.3 percentage points less likely to forego any expenses
The system’s part
Not isolated. The regression discontinuity isolates the effect of being eligible to borrow at the score cutoff. It does not compare the automated score with another way of setting eligibility. The authors describe the score as a complex formula using phone and M-PESA data and do not show that it learns from data.
Limits
Not yet checked.
Costs and harms
Not yet checked.
Later
Not yet checked.

Download data Cite Suggest a correction

Benefit type
Increased Inclusion
System
M-Shwari credit score (automated loan eligibility from mobile phone and M-PESA data)
Made by
Commercial Bank of Africa with Safaricom
Used by
Commercial Bank of Africa and Safaricom (M-Shwari on M-PESA)
Where
Kenya
Stage
In routine use
Date
Kind
Aggregate result
Broad benefit
Judging the Qualities of Objects, Services, or People, Making Decisions and Solving Problems
Last revised
31 October 2026

Sources

  1. Peer-reviewed study, by authors with no tie to the system.

    Suri T, Bharadwaj P, Jack W. Fintech and household resilience to shocks: Evidence from digital loans in Kenya. Journal of Development Economics 153 (2021): 102697. Abstract.

    The loans improve household resilience: households are 6.3 percentage points less likely to forego expenses due to negative shocks.

  2. Preprint, by authors with no tie to the system.

    Bharadwaj P, Jack W, Suri T. Fintech and Household Resilience to Shocks: Evidence from Digital Loans in Kenya. NBER Working Paper 25604, February 2019. The working paper version of the same study adds the design, the sample and the description of the credit score. The research was funded by Financial Sector Deepening Kenya.

    The chances that individuals can manipulate whether they fall on one side of the M-Shwari eligibility threshold is unlikely since the credit score is a complex formula using individuals' mobile phone and M-PESA data.

Details

AI system

Workflow

M-Shwari scored each customer at account opening from phone and M-PESA data and offered loans only above a fixed cutoff. The loans were small and short-term, with a 30-day term and a 7.5% monthly interest rate.

Follow-up

Conditions for replication

Customers need an M-PESA account and a score above the cutoff. The effect is measured for households near the cutoff, which the authors describe as a poorer part of the population.

Entry details

Sector

Consumer credit.

Purpose

Commercial Bank of Africa and Safaricom built M-Shwari's credit score to decide which customers may take digital loans.

Not yet checked: model, release, provider, role, configuration, inputs and outputs, autonomy, human oversight, AI contribution, beneficiaries, reach, distribution and access, duration, every measurement question, every money question, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.

Cite this record

Peter H. Vartanian. (2026). M-Shwari loans help Kenyan households weather shocks (AIBD-0880). Artificial Intelligence Benefits Database, edition of 31 October 2026. https://petervartanian.xyz/aibd/records/AIBD-0880.html (archived at https://doi.org/10.7910/DVN/AADSII, version 4.0)

BibTeX Data Field definitions