Portfolio Yield

Q: Why does Kiva choose to display the portfolio yield whenever possible, as opposed to interest rate?
A: To make sure that our lenders can understand the sort of interest rates and fees that borrowers are paying, whenever possible we display portfolio yield rate data on every loan profile.
Portfolio Yields

Portfolio Yields for our microfinance institution Field Partners are drawn from one of two sources depending on availability:

    Source of data Status of data
  MIX Market Financials shared by our partners with MIX Market  Often audited, but not always
  Kiva's own manual calculations Financials shared by our partners with Kiva Often audited, but not always

Using portfolio yield has several advantages: the MIX Market data usually comes from audited financials, and it takes into consideration the various ways that interest rates are calculated (flat vs. declining, annualized or not, etc.). It’s also largely available from a single source -- MIX Market -- which allows us to save our resources to invest in things like our ongoing efforts to use Kiva capital for driving innovative loan products like non-toxic lighting, clean water, and 0% loans through Kiva Zip.

Interest Rates

For our Microcredit Plus partners that are not microfinance institutions, portfolio yield data is rarely available. In these cases, we display the interest rates associated with the specific loan products that Kiva borrowers are being offered.

    Source of data Status of data
  Kiva's own manual calculations Self-reported data from non-MFI partners Not audited
For more information on Kiva's initiative to reduce interest rates and borrowing costs for the working poor, please see our interest rates Q&A.

Q: Is portfolio yield a good proxy for APR interest rates?
A: On average, we believe that the data indicates that it is. MIX Market performed an analysis that compared the two metrics, graphing them against each other to allow for a simple visual comparison:


source: MIX Market
Their conclusion was that, “Overall, this gives us a strong relationship between the two metrics -- yield and APR. The median difference in rates is 5 percent, at least with this first pass analysis. If we were to adjust for compounding and other factors the relationship would be even closer.”
It’s worth pointing out that while overall APR may be within a range of 5% (not 5 percentage points) higher than portfolio yield, there are also cases where the APR is below portfolio yield. As MIX Market explains, this can happen when an MFI has borrowers who aren’t paying on time. In those cases, the portfolio yield (which includes interest rates and fees, but only from borrowers paying on time) will be higher than the APR (which includes interest rates and fees charged -- but not necessarily paid -- by all borrowers).
We feel that a range of 5% is reasonable, especially considering the resources that would be necessary to collect loan schedules and fees for every loan. (To date, there have been more than 650,000 loans posted to Kiva).
Q: What would be required to collect individual APR interest rates and fees for each borrower on Kiva?
A: Microfinance Transparency explains the APR calculation here:
The true price of a loan includes not only interest but other charges required by the lender as well as other techniques that influence the amount of money the client actually has and the amount of time the client has use of that money. Because of these multiple factors, as well as differences in interest calculation methods, comparing the pricing of different loan products can be very challenging.
Here’s a breakdown of what’s needed to generate the APR for a specific loan:
  Loan amount   The amount being borrowed
  Loan term   How long the borrower can use the money
  Interest rate for loan   The rate at which interest is paid by a borrower (debtor) for the use of money that they borrow
  Method to calculate interest rate payments   Flat vs. declining, etc.
  Loan fees   Any upfront fees charged to the borrower
  Mandatory savings rates   Some banks require that borrowers save a percentage of their loan funds in their bank accounts during their loan cycle
  Savings interest   Interest paid to borrower on the balance of funds available in the borrower’s account

We currently collect some of this information, but would have to collect a few more data points to be able to calculate the APRs for individual loans (most importantly: loan fees, interest repayment schedules and mandatory savings rates). 
To give a sense of what would be required from an operational perspective, we should first explain that our Field Partners enter a significant amount of data for each loan that gets posted to Kiva, including the loan schedule, the loan use, borrower description, etc. To ensure reliability of this information, most data points that we collect from the partner have some level of validation associated with it.
So for each new data point that we add, we need to determine the operational cost of capturing it. On our end alone, we’d need to do things like define the process and business rules, designed in a way that is easy to use in low-technology and low-bandwidth environments, run quality assurance on the new functionality, message and train our 200+ partners, and update internal and partner-facing documentation. We also have to set up processes to screen for accidental erroneous data, which requires us to add validation processes for each field. 
That said, we are continuing to explore the relationship between APR and portfolio yield this year by capturing and calculating APR for a sample of loans on the website. Once we have that data in hand, we’ll be able to compare them to the portfolio yield data to see if they’re in-line. If the portfolio yields are not an effective proxy for APR as the data has so far suggested, then we’ll take that into account as we think about ways to ensure that the information on our website is as useful as possible.