About this pageGiveWell aims to find the best giving opportunities we can and recommend them to donors. We tend to put a lot of investigation into the organizations we find most promising, and de-prioritize others based on limited information. When we decide not to prioritize an organization, we try to create a brief writeup of our thoughts on that charity because we want to be as transparent as possible about our reasoning. The following write-up should be viewed in this context: it explains why we determined that we wouldn't be prioritizing the organization in question as a potential top charity. This write-up should not be taken as a "negative rating" of the charity. Rather, it is our attempt to be as clear as possible about the process by which we came to our top recommendations.
A note on this page's publication dateThe last time we examined CUMO was in 2010. In our latest open-ended review of charities, we determined that it was unlikely to meet our criteria based on our past examination of it, so we did not revisit it. We invite all charities that feel they meet our criteria to apply for consideration. The content we created in 2010 appears below. This content is likely to be no longer fully accurate, both with respect to what it says about CUMO and with respect to what it implies about our own views and positions. With that said, we do feel that the takeaways from this examination are sufficient not to prioritize re-opening our investigation of this organization at this time.
SummaryCUMO is a microfinance institution in Malawi owned by the U.K.-based charity Concern Universal.1 CUMO offers micro-loans and micro-insurance. It is not licensed to collect savings, though it encourages its clients to save through formal banks.2 Our review of CUMO consisted of reviewing information CUMO sent to us and phone conversations with Samson Hailu, the chair of CUMO's board of directors, and Ezikiel Phiri, CUMO's CEO. CUMO is notable for its commitment to monitoring and evaluation of its programs, including conducting regular surveys of clients who drop out of the program and commissioning the rating agency M-CRIL to conduct a detailed social performance assessment. We do not currently recommend CUMO, feeling that the Small Enterprise Foundation provides a better opportunity for donors interested in supporting microfinance.
Table of Contents
Key questions about social impactIn evaluating CUMO, we asked our key questions for evaluating microlending charities. The sections below cover its answers to our key questions (see the previous link for why we consider these questions important).
Is the organization focused on social impact?
How frequently do borrowers drop out of the program?CUMO reports drop out rates of 5.5% in 2008 and 1.1% in 2009.3 CUMO calculates its drop out rate as the proportion of clients who dropped out during out during the period over the number of clients that existed at the beginning of the period,4 and told us that a drop out is any client who completes a loan cycle and does not take out another loan right away.5 We have remaining questions about apparent inconsistencies in CUMO's reported drop out rates.6 CUMO told us that it considers its retention rate to be one of the most important indicators of its success and that CUMO's board sets targets for this rate.7 Loan officers' pay is partially tied to the drop out rate among clients they serve.8 For more information on microfinance drop out rates and how they are calculated, see our microfinance glossary.
Does CUMO monitor why borrowers drop out?M-CRIL notes that CUMO conducted studies "of good quality" in 2005, 2006, 2007, and 2008 of the reasons why clients drop out of the program. CUMO shared with us reports from surveys conducted in 2003 and 2008. The 291 drop outs surveyed in 2008 reported the following reasons for leaving:9
- Repayment frequency (30.1% of sample)
- Expensive interest and fees (21.6%)
- Short repayment length (17.0%)
- Loan officers' behavior (11.5%)
- Business seasonality (8.0%)
- Family crisis (5.2%)
- Unwilling to borrow (5.1%)
- Failure to repay (4.0%)
- Reallocation (3.1%)
- Small loan size (1.5%)
Does CUMO prevent client over-indebtedness?CUMO told us that it does the following to prevent over-indebtedness among clients:14
- Conducts regular visits to client businesses.
- Tries to only provide loans for which installments do not exceed 40% of the profits generated by the client's business over the installment period.
- Requires group members to approve loans before disbursement. Since group members are required to cover any delinquent payments, it is expected that they will not approve loans that a member will have trouble repaying.
Are borrowers protected against harassment by loan officers and group members?The rating agency M-CRIL reports that CUMO's internal auditors visit each of CUMO's 11 districts at least annually and evaluate "measures taken in the case of delinquent loans to make sure that staff are acting in accordance with CUMO policies and clients are not being unduly harassed." M-CRIL also reports that "audit reports are detailed and of good quality, with recommendations for corrective action."17 M-CRIL also notes that "regional managers and BAS [business area supervisors] provide their mobile numbers to the clients so that in the event of any problem the clients can freely report to CUMO,"18 though M-CRIL still believes that "a formal grievance addressal process should be made available to the clients. Most clients depend on the FSOs [financial service officers] for conveying any form of feedback."19 Loan officers' pay is partially based on their performance, specifically managing a large number of clients and a high portfolio growth rate, and maintaining a low drop out and a high repayment rate.20 It is possible that incentives to have a large portfolio and high repayment rate could lead to loan officers pressuring clients to repay or to take out loans they wouldn't otherwise have taken out. On the other hand, loan officers are also rewarded for maintaining a high client retention rate, which could lead to better treatment of clients. We note that CUMO planned to intensify arrears collections in May and June 2010.21 We would like to know how CUMO protects clients from harassment during such activities. CUMO told us that pressure is applied by group members, not loan officers, but after discussing this issue with CUMO,22 it remains unclear to us how clients are persuaded to repay during such activities.
What interest rates does CUMO charge?CUMO provided example repayment schedules for its 3 most popular loan types (out of 4 total types). We used these schedules to calculate the cost of borrowing to clients.23
|Loan type||% of CUMO's loan distributions||Monthly rate||Annual percentage rate (APR)||Effective interest rate (EIR)|
|Masika (business loan targeting small scale entrepreneurs)||64%||13%||156%||354%|
|Fumba (agricultural input loan for existing Masika borrowers)||31%||6%||72%||102%|
|Mtenthandevu (small holder tea growers product)||3%||4%||53%||69%|
|Kasupe (business loan targeting medium scale entrepreneurs)||2%||N/A||N/A||N/A|
What is CUMO's repayment rate?CUMO appears to have a fairly low repayment rate in recent quarters.26 CUMO calculates its repayment rate as 100% minus the portfolio at risk greater than 14 days ratio. CUMO told us that it does not reschedule loans.27 CUMO told us that decreases in the repayment rate have generally been due to crop failures and unexpected policy changes affecting farmers' ability to sell their products.28 CUMO also told us that it calculates the following collection/repayment rates:29
- Total repayments in the period less prepayments divided by installments and amounts due including past dues. CUMO reported this rate as 87% as of June 2010.
- Repayments in the period less prepayments divided by amounts falling due in the period. CUMO reported this rate as 93% as of June 2010.
What are CUMO's clients' standards of living?According to a 2004 study based on a randomly-selected client sample,30 3% of CUMO's clients were "non-poor," 35% "upper poor," 50% "middle poor," and 13% "poorest."31 It is not clear to us how these categories compare to internationally-recognized poverty lines. CUMO describes the "middle poor" (50% of CUMO's clients) as:32
The middle poor have about 1.5 acres of land. They have some livestock, mainly chickens, and slightly better housing than the poorest, usually a rectangular thatched mud house. They have food for about eight months and this is often stocked in a small granary. They have better but few items of clothing. They also have more, but basic, household items like plates, pots and small chairs. The middle poor send their children to school albeit with poor clothing.In 2009, the rating agency M-CRIL surveyed 138 new clients using the Progress Out of Poverty Index for Malawi.33 Based on this survey, M-CRIL estimated that 58.7% of new clients live on less than $2 per day at purchasing power parity.34 Other indicators of standard of living from the survey include: 85% of new clients use an unprotected water source, 69% have no furniture in their home, and 42% own a bicycle.35
Monitoring and evaluationAs discussed above, CUMO conducts annual surveys of drop out clients to determine why they have left the program. We have seen reports from 2003 and 2008, and the rating agency M-CRIL reports reviewing reports from 2005-2007. In addition, in 2009 CUMO commissioned the rating agency M-CRIL to conduct a detailed social performance assessment of CUMO, including evaluations of its focus on the poor, its client protection practices, and results of its satisfaction and drop out reasons surveys.36 CUMO has allowed us to post this assessment publicly.
Room for more fundsThe chair of CUMO's board of directors told us that additional donations would be used to expand lending activities to additional areas of Malawi and to pilot new products.37 CUMO's CEO told us that such new products might include loans with longer repayment periods and support for social enterprises such as provision of safe drinking water.38 We do not know how much in additional donations CUMO could productively use for these projects in the next year. The rating agency M-CRIL, in its 2009 assessment of CUMO, noted that "CUMO is in the unique position of having surplus funds at its disposal."39 CUMO told us that it is not in need of funding for its established loan products, though its reserves are lower than they were at the time of the M-CRIL assessment.40
- Drop out. Can CUMO explain the apparent inconsistencies in its drop out data?
- Client protection. Can CUMO share reports from its audit process?
- Room for more funds. How much in additional donations could CUMO productively use in the next year?
- Concern Universal. Development projects in Malawi. http://www.concernuniversal.org/index.php?/where_we_work/malawi/projects (accessed May 18, 2010). Archived by WebCite® at http://www.webcitation.org/5polFOQbx.
- CUMO. Baseline impact assessment survey and analysis of clients (2004) (DOC).
- CUMO. Impact assessment: Executive summary (2006) (DOC).
- CUMO. Performance indicators (2008 Q1). CUMO asked that we keep this document confidential.
- CUMO. Performance indicators (2008 Q2). CUMO asked that we keep this document confidential.
- CUMO. Performance indicators (2008 Q3). CUMO asked that we keep this document confidential.
- CUMO. Performance indicators (2008 Q4). CUMO asked that we keep this document confidential.
- CUMO. Performance indicators (2009 Q2). CUMO asked that we keep this document confidential.
- CUMO. Performance indicators (2009 Q3). CUMO asked that we keep this document confidential.
- CUMO. Performance indicators (2009 Q4). CUMO asked that we keep this document confidential.
- CUMO. Performance indicators (2010 Q1). CUMO asked that we keep this document confidential.
- CUMO. Performance indicators (2010 Q2). CUMO asked that we keep this document confidential.
- CUMO. Response to GiveWell report (DOC).
- CUMO. Social performance standards report (2008) (XLS).
- GiveWell. Questions for microfinance charities.
- GiveWell. CUMO interest rate calculation (XLS).
- Hailu, Samson. CUMO Board Chair. Phone conversation with GiveWell, June 1, 2010.
- Langeman, Tim. Calculating interest rates with excel (accessed April 27, 2010). MFTransparency blog, February 24, 2010. Archived by WebCite® at http://www.webcitation.org/5pJ4ffoh6.
- M-CRIL. Social performance assessment and institutional case study of CUMO, Malawi (PDF).
- Mix Market. Data on CUMO (XLS).
- Mix Market. Glossary. http://www.mixmarket.org/en/glossary (accessed May 18, 2010). Archived by WebCite® at http://www.webcitation.org/5pomL0lun.
- Phiri, Ezikiel. CUMO CEO. Email to GiveWell (DOC), June 22, 2010.
- Phiri, Ezikiel. CUMO CEO. Phone conversation with GiveWell (DOC), August 5, 2010.
- Reserve Bank of Malawi. Interest rates. http://www.rbm.mw/data/findata/index.asp?content=rbmir (accessed May 4, 2010). Archived by WebCite® at http://www.webcitation.org/5pTn2susZ.
- 1. Concern Universal, "Development Projects in Malawi."
- 2. "As a private company, CUMO does not collect deposits but it has active strategic partnership with a number of banks, particularly Standard Bank in which its clients are encouraged to save in group accounts, and internally, for on-lending." M-CRIL, "Social Performance Assessment and Institutional Case Study of CUMO, Malawi," Pg 6.
- 3. CUMO, "Performance Indicators (2008 Q1-2010 Q1)."
- 4. "The client drop-out rate is calculated by counting the number of clients that dropped out during the period, dividing the result by number of clients the company had at the beginning of the period in question and multiplying the result by 100 to get a percentage." Ezikiel Phiri, email to GiveWell, June 22, 2010.
- 5. Ezikiel Phiri, phone conversation with GiveWell, August 5, 2010.
- 6. For example, CUMO quotes its drop out rate for 2009 as 1.1% (CUMO, "Performance Indicators (2010 Q1)," Pg 6) but quotes its drop out rate for the each of the first three quarters of 2009 as 2.3%, 8.7% and 1.1%, respectively (we are missing data from the fourth quarter). Given CUMO's formula for the drop out rate, we would expect that the drop out rate for any period would be greater than the rate for any part of that period, and certainly for any shorter period with the same start date (the denominator will be the same but the numerator, i.e. the number of drop outs, will be less than or equal for the shorter period).
- 7. Samson Hailu, phone conversation with GiveWell, June 1, 2010.
- 8. M-CRIL, "Social Performance Assessment and Institutional Case Study of CUMO, Malawi," Pg 13.
- 9. CUMO, "Exit Survey Report (2008)," Pg 5.
- 10. "Some of the things that clients liked least about the program were:...Being forced to repay even when someone is in experiencing a crisis like death." CUMO, "Exit Survey Report (2003)," Pg 2.
- 11. "A mandatory "funeral benefit insurance" scheme operates for all CUMO loans, paid for by a 2% fee levied on clients at the time of loan application. This covers the full amount of the outstanding loan in the event of death of the member plus a payout of MWK 5,000 for funeral expenses. The insurance safeguards the group from having to discharge loans taken out by a deceased member." M-CRIL, "Social Performance Assessment and Institutional Case Study of CUMO, Malawi," Pg 16.
- 12. "Some of the things that clients liked least about the program were:...The two weekly repayment schedule." CUMO, "Exit Survey Report (2003)," Pg 2.
- 13. CUMO, "Response to GiveWell Report," Pg 1.
- 14. CUMO, "Responses to GiveWell Report," Pg 4.
- 15. Samson Hailu, phone conversation with GiveWell, June 1, 2010.
- 16. CUMO, "Impact Assessment: Executive Summary (2006)," Pg 2.
- 17. M-CRIL, "Social Performance Assessment and Institutional Case Study of CUMO, Malawi," Pgs 15-16.
- 18. M-CRIL, "Social Performance Assessment and Institutional Case Study of CUMO, Malawi," Pg 18.
- 19. M-CRIL, "Social Performance Assessment and Institutional Case Study of CUMO, Malawi," Pg 24.
- 20. M-CRIL, "Social Performance Assessment and Institutional Case Study of CUMO, Malawi," Pg 13.
- 21. "Management is working to see the PAR situation reduced by end of June, 2010. In this regard, Management has initiated an intensified arrears collection exercise that also targets the harvesting period and is expected to peak in the months of May and June. The effort has already started yielding increased collections from the central region especially the Kasumbu area." CUMO, "Performance Indicators (2010 Q1)," Pg 7. CUMO told us that it was primarily looking to collect arrears from cotton growers whose supply chain was temporarily interrupted by a decision by the president of Malawi to institute a price floor on cotton. The main cotton buyer left the country as a result and was replaced some time later by a state-controlled buyer. Samson Hailu, phone conversation with GiveWell, June 1, 2010.
- 22. Ezikiel Phiri, phone conversation with GiveWell, August 5, 2010.
- 23. Data from CUMO, "Response to GiveWell Report," except for data on "% of loans disbursements," which is from CUMO, "Performance Indicators (2010 Q1)," Pg 5. Annual percentage rate (APR), effective interest rate (EIR) and monthly rate were calculated by GiveWell. Note that the APR is equal to 12 times the monthly interest rate, while the EIR fully incorporates "compounding," whose relevance to microloans is debatable. Calculation in GiveWell, "CUMO Interest Rate Calculation." For more information on calculating interest rates for microfinance banks, see Langeman 2010.
- 24. Fees and savings requirements described in Ezikiel Phiri, email to GiveWell, June 22, 2010.
- 25. Reserve Bank of Malawi, "Interest Rates."
- 26. "On time repayment rate": 2008 Q1: 96.50% 2008 Q2: 96.00% 2008 Q3: 95.60% 2008 Q4: 96.90% 2009 Q1: 93.50% 2009 Q2: 93.80% 2009 Q3: 85.41% 2009 Q4: Quarterly rate not listed 2010 Q1: 85.21% 2010 Q2: 87.24% Data from CUMO, "Performance Indicators (2008 Q1-2010 Q2)."
- 27. Ezikiel Phiri, phone conversation with GiveWell, August 5, 2010.
- 28. Samson Hailu, phone conversation with GiveWell, June 1, 2010.
- 29. CUMO, "Response to GiveWell Report," Pg 1. CUMO notes these rates do not take into account repayments received by the company that could not be allocated by the close of the reporting period.
- 30. "A total of 325 clients within 25 client groups took part in this assessment...Groups were selected using a stratified simple random sampling method (described in more detail in the strategy paper) from all four of the areas in which we work." CUMO, "Baseline Impact Assessment Survey and Analysis of Clients (2004)," Pg 5.
- 31. CUMO, "Baseline Impact Assessment Survey and Analysis of Clients (2004)," Pg 6.
- 32. CUMO, "Baseline Impact Assessment Survey and Analysis of Clients (2004)," Pg 18.
- 33. "Client level data presented in the following sections of the report is drawn from CUMO research studies supplemented by a client survey of 138 new (less than a year with CUMO) clients designed by M-CRIL and conducted by CUMO staff...The PPI Malawi (Feb 2009) was used in the client survey (138 clients) and the results are presented below." M-CRIL, "Social Performance Assessment and Institutional Case Study of CUMO, Malawi," Pgs 22-23.
- 34. "The data suggests that CUMO has a reasonable poverty outreach, with 58.7% of the sample at entry living below the $2 poverty line at purchasing power parity (PPP), and 36% below the national poverty line. However, it is clear that CUMO is not reaching the poorest, since the sample statistics reveal that outreach is not even as much as the actual presence of poor in the population for food and national poverty lines. 8.2% of its clients can be classified as non poor (above $2.5), and this indicates that while CUMO has designed its operations to include only poor, without client targeting strategies some non-poor are included in the programme." M-CRIL, "Social Performance Assessment and Institutional Case Study of CUMO, Malawi," Pg 23.
- 35. M-CRIL, "Social Performance Assessment and Institutional Case Study of CUMO, Malawi," Pg 24.
- 36. M-CRIL, "Social Performance Assessment and Institutional Case Study of CUMO, Malawi."
- 37. Samson Hailu, phone conversation with GiveWell, June 1, 2010.
- 38. Ezikiel Phiri, phone conversation with GiveWell, August 5, 2010.
- 39. M-CRIL, "Social Performance Assessment and Institutional Case Study of CUMO, Malawi," Pg 10.
- 40. Ezikiel Phiri, phone conversation with GiveWell, August 5, 2010.