The GiveWell Blog

Microfinance interest rates

One of the more difficult things to understand about the microfinance institutions we’ve investigated is the “true” interest rate they’re charging their borrowers. In July 2009, David Roodman of the Center for Global Development wrote:

It appears that many MFIs impose subtle fees that effectively raise interest rates. Some charge one-time loan origination fees. Some require borrowers to deposit a percentage of each loan amount with the MFI in a savings account that pays interest at a rate lower than that on the loan. Some overcharge for credit-life insurance bundled with the loan. Another criticized practice is to charge interest on the full loan amount even as the outstanding balance declines over the repayment cycle. Such “flat-rate” interest effectively doubles the interest rate compared to “declining-balance” interest since the average balance over the cycle is half the starting amount. Also, MFIs may also prefer to quote their rates on a monthly basis, hoping to exploit borrowers’ ignorance of how a seemingly modest 6 percent per month compounds into 100 percent per year.

Our experience with the microfinance organizations we have investigated to date suggests that these are real issues that donors should be aware of when interpreting interest rate data.

To be clear, we don’t think charging a high interest rate signifies wrongdoing on the part of a microfinance bank. High interest rates may be the best way to minimize losses and serve more people, and client participation at high interest rates may be an indicator that they are getting a service they value. We just want to note how striking the difference is between the initial “cited” interest rate donors often hear about and the “effective” interest rate taking all factors into account.

We recently evaluated a microfinance institution (MFI) in Malawi, the Microloan Foundation, as part of our process for distributing a grant to an economic empowerment charity in Sub-Saharan Africa. Its stated interest rate for its most popular loan type is 20%, but:

  • 20% is the rate over the course of the 4-month loan. The “nominal Annual Percentage Rate (APR)” (a common way of stating interest rates in standardized terms and the rate which U.S. lenders are required to provide to borrowers) of this loan, with no other costs, would therefore be 60%.
  • Interest is calculated as a flat rate. 20% of the whole loan amount is charged each payment, instead of 20% of the remaining loan balance. This method raises the nominal APR from 60% to 93%. On a $100 loan of this type, a borrower would pay $20 in interest compared with only $12.80 on a loan with declining balance interest.
  • Payments are due every two weeks, instead of every month, so that the first payment is due only two weeks after the loan is made. Requiring 8 payments instead of 4 raises the nominal APR from 93% to 102%.
  • The Microloan Foundation requires borrowers to hold 20% of the loan amount in a savings account which cannot be accessed until the loan is repaid. On a loan of $100, this requirement reduces the effective size of the loan to $80, while decreasing the effective size of the last payment due by $20 (because savings are then accessible). The savings requirement raises the APR from 102% to 149%.
  • Arguably, the APR used above (i.e. the “nominal APR”) understates the interest rate people are paying because it does not take into account the compound value of interest. (Wikipedia’s entry on APR has a discussion of the relevant issues.) At relatively low interest rates, such as the interest rates we’re used to in the U.S., the “nominal APR” (what is usually reported) and the “effective APR” (the “mathematically true” interest rate) are usually very close to each other – but at the much higher interest rates charged by microfinance institutions, the “effective APR” can be considerably higher, raising the question of which one should be quoted to give Americans the best picture of what people are being charged. The Microloan Foundation’s nominal APR of 149% is equivalent to an effective APR of 326%.

Not only are these final “effective” interest rates many times bigger than the initial “20%” figure, they’re also significantly higher than would be implied by looking at MLF’s nominal gross portfolio yield according to MixMarket (93% for 2008; Mix Market defines this as interest and fees divided by the gross average loan portfolio ).

MLF doesn’t charge fees on its loans, but other microfinance institutions do, and these can cause further significant increases in the effective interest rate. (For example, adding a 5% fee to the beginning of a loan that without fees would have a nominal APR of 40% raises its nominal APR to 66%.)

From what we’ve seen, fees, compulsory savings, and the flat interest rate method seem to be fairly common among microfinance institutions. Of the 65 MFIs who had submitted a Social Performance Standards Report to Mix Market as of late 2009, 42% use the flat interest rate exclusively, 29% held compulsory savings accounts, and 72% collected fees on at least some loan products (according to their Mix Market profiles).

Understanding the true cost of credit for a borrower is important for reasons discussed previously:

  • High interest rates (combined with high rates of repayment and low drop out rates) show that borrowers are willing to pay a high price for the loan, arguably implying that they value this service.
  • If interest rates are low, microfinance institutions may be effectively giving cash transfers, at which point the mere fact of participation becomes less meaningful, and it becomes more important to ask whom the handouts are going to.
  • On the other hand, the higher the interest rate, the more we worry about whether borrowing is good for the borrowers (and about anecdotes like this one).

A fairly new initiative (started in 2008), MFTransparency, is working to create a more open discussion about microloan pricing, and has compiled and published interest rate data for two countries, Bosnia and Cambodia. We look forward to following the progress of this initiative and to drawing on its data to inform our investigations of microfinance organizations.


  • Basti Schwiecker on April 2, 2010 at 3:33 pm said:

    Interesting read, especially since I wrote my master thesis about exactly the same topic (back in 2005:

  • Deniz Hassan on April 9, 2010 at 11:26 am said:

    At the MicroLoan Foundation, we believe transparency for both our clients and donors is absolutely key and respect that the article is looking to address this. Having been singled out as an MFI that is misleading its donors by charging a much higher interest rate than it states we feel compelled to add to the article to hopefully give a bit more balance to what the MicroLoan Foundation do (including our interest rates) and why.

    To look at microfinance trends broadly across the world, comparing one “MFI” working within one region of the world to another working in a different region is intrinsically misleading.. The key part of the term microfinance is ‘micro’ – this means looking at each MFI in respect to its beneficiaries country’s economies and the actual demographic of the beneficiaries.

    In other pages of the GiveWell site GiveWell criticise MicroLoan Foundation for only providing information back to 2006 or 2007, whilst other organisations provide information stretching much further back. There is a simple reason for these differences, MicroLoan Foundation is a much younger organisation than those to whom GiveWell seek to compare it to (started in 2002 in one of the poorest regions of Malawi with one man, a bike and $15,000), it has never received any major one off donation from any organisation and has built itself up incrementally valuing every $ donated – at our stage of development we feel our priority should be to support the thousands of needy people in Malawi with small loans and business training, rather than spend thousands of dollars on a sophisticated MIS

    At MicroLoan Foundation we work primarily in Malawi, one of the 10 poorest countries in the world the countries infrastructure is limited, the majority of people live subsistence livelihoods and the level of education is low. We focus our work in the rural areas with people who literally have next to nothing and have never previously had access to financial services and up to 50% of whom are illiterate. In addition we only make loans to poor women who want to start or develop existing businesses. We do not make high value loans to salaried employees, favoured by the MFI’s to drive sustainability.

    We don’t simply see ourselves as a traditional “MFI”; we insist that each loan group receives 8 modules of training before they receive their loans, we equip our staff with motorbikes so that they are able to travel significant distances from population centres to visit our client groups (who otherwise would not be aware of our existence nor be able to afford to travel to our branch offices). We could break out the cost of the service that we provide between signing on, education and training and interest, but we choose not to for the ease of understanding for our clients. Our interest rates are in place to create financial sustainability i.e. to cover the cost of providing the above service and are calculated in accordance with CGAP’s standard method for interest rate setting. Whilst experience in Malawi would concur with the findings of Schmidt and Tschach (2001) and Rosenberg (2002) referenced in the above posters paper – that our clients are able to generate profits of many times the amount that we loan to them – we would posit that there is likely a causal link between the level of development of the community and the rates of profit that its micro entrepreneurs are able to yield from micro finance.

    Givewell’s calculation of MicroLoan Foundation’s rates is based on the assumption that clients pay back exactly on time every time, however elsewhere on their site they have questioned our repayment rate, despite us explaining our repayment rate to them on several occasions. In the last 3 years 98% of our clients have paid back the money that they have borrowed from us. However a small % of our clients take longer than the agreed repayment terms. than the agreed period to repay.. Rather than penalise the clients with fines, levies an additional interest for late repayments we take a more understanding approach and allow them flexibility to repay. Their failure is our failure. We currently do not undertake complex tracking of this in our MIS, we keep things simple and allow these clients to simply become clients in arrears, who experience tells us will repay when they are able to. Applying the Givewell interest calculation in these circumstances would see the effectve/annualised rates fall significantly, probably somewhere in line with the Mix market rates they refer to in this article. We have more important things to do than to spend our time calculating this exact figure under our flexed repayment terms – REMOVE

    But of course it is not the input to the micro finance borrowers that are important, it is the outcomes of those loans on the clients and their families. In our ongoing dialogue with Givewell we have provided the evidence of detailed studies to show our social impact in Malawi. This is coupled with case studies and frequent visits from donors and donor groups has confirmed to us that the work we do is having a positive impact on the communities in which we operate. Having recently undertaken successful pilot studies to try to standardise our social impact reporting, we are now ready to start investing in this in the knowledge that our funds are being spent in a way which will further our aims.

    There is a natural life cycle to micro finance organisations as they grow in sophistication which is also dependent on the communities in which they’re based. Microfinance in more developed countries and the higher valued salary loans will inevitably have a lower interest rate, be able to reach scale quicker and be able to produce more sophisticated documentation to justify its existence. To look at these in isolation without taking the conditions above into consideration risks donors opting to support microfinance projects that are the most sophisticated, not necessarily those supporting the most needy.

    Givewell’s interest rate calculations generate such high rates of interest for MicroLoan because they deduct the moneys that we facilitate clients to save with local Malawian banks from the amount that they borrow from us. We understand the logic behind this approach but do not think that the outcome it generates is a fair reflection on us. We require that clients save a set amount of their loan values with local banks, partially as collateral against default (our high repayment rate means instances of this are limited) but also to introduce them to the formal banking system and to provide them with a savings facility

    We acknowledge it is our challenge for the coming year to prove that the work we do reaches the most needy and have a 2 year project in hand to prove it. it would be good to hear a similar commitment from GiveWell in which they take into account nuances between different developing country situations and the stage of the MFI in their lifecycle.. We are always available for comment and questions and seek to be 100% transparent to donors and beneficiaries.

  • Holden on April 14, 2010 at 7:24 pm said:

    Deniz, thanks for the comments. A couple of points of clarification:

    • We have not accused MLF of misleading clients.
    • We do not believe it is wrong for MLF to charge high rates of interest (and our post above states this).
    • We have a high opinion of MLF overall and find its transparency and explicit focus on social impact (not just financial sustainability) to be superior to that of most other MFI’s we have seen.
    • That said, as you acknowledge there are still gaps in the information needed by an outsider to assess MLF’s likely social impact. That’s why we don’t currently recommend MLF.

    GiveWell’s goal is to recommend demonstrably great charities to individual donors; a lack of recommendation is different from a positive claim that good isn’t being accomplished.

  • james on April 16, 2010 at 9:10 am said:

    I, like many others, have given money on kiva but then I realized there was a better way. Microfinance is to financially support entrepreneurs (they may receive other support such as bookkeeping from the loan sharks… err, lenders but I do not know for sure). There are entrepreneurs already established and selling their products, likely locally, and certainly internationally in the fair trade environment. Loans can help but repaying them with 100 percent interest in six months is possibly not the best solution to ending poverty. The best way to support a business is to buy their products.

Comments are closed.