Dr. P.V. Viswanath



Economics/Finance on the Web
Student Interest

  Home/ Microfinance/ Exams/  

Spring 2013




  • If your answers are not legible or are otherwise difficult to follow, I reserve the right not to give you any points.
  • If you cheat in any way, I reserve the right to give you no points for the exam, and to give you a failing grade for the course.
  • You may bring in sheets with formulas, but no worked-out examples, or definitions, or anything else.
  • You must explain all your answers.
  1. (10 points) What is the difference between direct and indirect MF securitizations?  In what ways is this difference important?  (Provide at least two implications.)
  2. (10 points) One of the complaints against the Banco Compartamos IPO was it enriched the equity investors at the expense of the borrowers who were charged high interest rates, which allowed Banco Compartamos to increase their retained earnings.  Richard Rosenberg provides a defense against this charge in the following words:

    From the very inception of the microcredit program, the leaders of Compartamos stayed focused on their vision of reaching a million poor Mexican women with loans, as quickly as possible.  It seems reasonably clear that, in the absence of high retained profits, Compartamos’ expansion from 1995 to 2000 would have been considerably lower.

    Can you suggest how Compartamos might have increased their retained earnings and still not have enriched shareholders at the expense of borrowers? 
  3. (20 points) Consider a village where all households are eligible for a loan from a microfinance enterprise.  Suppose that half of those households borrow from a microfinance enterprise, and that half of them do not borrow at all.  The total number of children of participant borrower households is 119 and the number of nonparticipants is 143.  Before borrowing from a microfinance enterprise, the number of participant borrowers’ children enrolled at school was 51, and that of nonparticipants was 71.  After joining the microfinance program, the number of children in school of program participants increased to 65, which in turn made the nonparticipants increasingly inclined to send their children to school.  Suppose that, on average, for every two additional children that participants in the microfinance program send to school, there is a spillover effect of one additional child from the nonparticipant group who will go to school.  Compute the percentage of children that go to school in both the participant and nonparticipant groups once the microfinance program has been set up, assuming that the birth rate in the village throughout the duration of the program is 5 percent.  Then evaluate the merits of the following statement: “Microfinance has no effect on education,” and explain your answer.
  4. A bank is involved in a microloan program in a village.  The bank has done some market research and feels comfortable in dividing the set of potential entrepreneur/borrowers into three types.  All three types of entrepreneurs can be considered risk-neutral; the bank itself considers itself to be risk-neutral, as well, in terms of its lending decisions, at least in terms of this particular microfinance venture.    
    Here is some additional information that the bank has regarding the three types.  If a type 1 entrepreneur invests $200, she gets a gross return of $300 with certainty.  If a type 2 entrepreneur invests $100, she gets $400 with certainty.  And finally, if a type 3 entrepreneur invests $100, she gets $300 with probability 0.8 and $0 with probability 0.2.  The bank can determine if potential borrowers are of type 1 (henceforth, high-type borrowers), but it can’t distinguish between entrepreneurs of type 2 and 3 (henceforth, low-type borrowers), but it does know that 40% of the low-type borrowers are of type 2, and the other 60% are of type 3.  All borrowers, on the other hand, can recognize each other’s types.  Under these conditions, the bank decides to extend individual loans to high-type borrowers and group loans with joint liability to low-type borrowers.  As a result, a low-type borrower may have to repay for a defaulting peer.  The administrative expenditure associated with lending to high-types is $20, while the administrative expenditure associated with lending to low-types is $30 (because the bank has to put in additional time and effort to ensure that groups are formed and to enforce debt repayments).  In addition, the bank has to pay a 12 percent cost for the use of the funds that it uses in this microfinance lending program.  Assume that the borrowers are protected limited liability.
    1. (15 points) You are a consultant hired by the bank to help it decide on the right interest rates to charge.  If the bank only aims to break even, calculate the interest rates charged to high types and low types.
    2. (5 points) After you have computed your answers in part a., the bank’s research department comes back to you to inform you that they have made a mistake – in fact, type 3 entrepreneurs will only make $280 from their investments when they are successful.  Would you increase the interest rate charged to the low-type borrowers?  Why?
  5. (30 points) Answer any three from these questions (no more than five sentences for each answer):
    1. Suppose you are a management consultant for an MFI. You are worried that families who chose to accept microloans from your client might have been more entrepreneurial than families that didn't. Explain how this might bias your results.
    2. Some people, such as Bhaduri (1973) promote microfinance as a solution to the high interest rates charged by moneylenders because of their monopoly power. According to this, there may be no market failure. How would your strategy to introduce microfinance differ if you believed the monopoly power theory as opposed to the market failure theory?
    3. What are the key elements of a credit cooperative? How is a credit cooperative different from a ROSCA?
    4. Why might competition not be a good thing in microfinance, even from the viewpoint of borrowers?
    5. What are the advantages of insuring farmers against bad weather rather than against bad harvests? When would such a plan work well and when wouldn't it?
    6. Could microfinance entrench gender roles? Explain.
  6. (20 points) Answer any two of these questions (no more than five sentences for each answer):
    1. What are some of the social factors that make it easier or more difficult for people to manage their money?
    2. Why might an auction ROSCA be considered an efficient intermediation system?
    3. Why is health insurance much more difficult to provide than funeral insurance?
    4. Why might the stated interest rate and the effective rate be different on loans given by moneylenders?

Midterm Solutions

  1. In direct securitizations, microloans originated by MFIs are pooled into an SPV (special purpose vehicle) that is typically a trust. In indirect securitizations, loans made to MFIs get pooled into an SPV. Since the MFIs themselves depend upon the loans that they make, an indirect securitization also depends ultimately upon the microfinance loans originated by the MFIs. However, in an indirect securitization, only the net cashflows of the MFI are accessible to the securityholder. On the other hand, the MFI might have equity that could be claimed by the securityholders; the microborrowers are unlikely to have much collateral or equity that could be accessed by the securityholders. There is likely to be much more information about the MFI cashflows, which are the basis for an indirect securitization, as compared to the microborrower cashflow information.
  2. Compartamos could have provided a discount to its borrowers in the form of equity. Thus, the high interest rates would have been in the nature of a tied sale -- a microloan plus an equity purchase. This would have allowed the borrowers to reap the eventual higher value of the equity; this is similar to the Grameen Bank, where all the borrowers are also stockholders. Another possibility would have been a tied loan to Compartamos that would be used to finance growth in loans.

  3.   Numbers Proportions   Numbers Proportions
    Participants 119 Nonparticipants 143
    Initial # Schoolgoers 51 0.428571 71 0.496503
    Spillover 7
    Final # Schoolgoers 65 0.546218 Final # Schoolgoers 78 0.545455
    Natural increase 2.55 Natural Increase 3.55
    62.45 0.52479 74.45 0.520629
    Diff w/o spillover adjustment   0.096218 Diff w/o spillover adjustment   0.024126
    Diff of Diffs w/o spillover adj   0.072093
    Diff w/ spillover adj   0.096218 Diff w/ spillover adj   -0.02483
    Diff of Diffs w/ spillover adj   0.121044

    Because of the spillover, the number of schoolgoing children in the control group increases by 7 = 0.5(65-51). Furthermore, the population grows by 5% and assuming that the growth rate in both participant and non-participant groups will also be 5% even without the microfinance treatment, we'd have a natural increase of 51(0.05) or 2.55 schoolchildren in the participant group and 71(0.05) or 3.55 in the nonparticipant group.
    Using the difference of difference method, we have a treatment impact of 9.62%, if we ignore the fact of the spillover impact on the schoolgoing propensities of the nonparticipant group. If we take that into account, as well, we have a differential treatment impact of 12.1%. It should also be noted that if we simply looked at the schoolgoing proportions in the treated and non-treated groups, there would hardly be any difference -- 52.48% for the treated and 52.06 for the non-treated, and we might conclude that the program did not make a difference.
    Of course, this assumes that the nonparticipant group is probabilistically similar to the participant group. If the factors that led the participant group to accept the microloans might also have led them to a greater propensity to send their children to school during the period of the treatment, then our numbers might be biased. However, while we might suspect that there is a selection bias that would lead to families with more education (and higher prospensities to school their children) to take loans, it's difficult to see a selection bias that affects the growth rate. In fact, the proportion of initial schoolgoers in the participant population is lower than in the non-participant population.
    Note: Of course, we have not conducted any kind of statistical testing, which is necessary to make final conclusions.
    1. Type 1 entrepreneurs get a return of 1.5 per dollar invested without any risk ($300 for an investment of $200). Furthermore, the bank can identify type 1 investors. The administrative costs of lending to these entrepreneurs is $20, or 10% per dollar lent; in addition, there is a cost of funds of 12 percent, for a total of 22%, i.e. R = 1.22.
      For the low-types, since the borrowers know each other's types, they will engage in assortative matching. The type 2 groups will also repay their loans, since their investments are risk-free. To begin with, let us assume that if there is only one successful entrepreneur in the type 3 groups, s/he will be able to repay the obligations of both the people in the group. In this case, thee type 3 groups will repay fully with probability 1-(0.2)2= 0.96. Hence the average repayment rate will be 0.4+(0.6)(0.96) = 0.976. The total costs will be 30/100 + 12% or 42%; hence the break-even value of R will have to be 1.42 = 0.976R, or R = 1.4549. With this interest rate, we can confirm that even if there is only one successful entrepreneur in the type 3 groups, s/he will be able to repay the obligations of both group members, since 300 > 2(100)(1.4549) = 290.98.
    2. However, if type 3 entrepreneurs will only make $280 from their investments if they are successful, then there will not be full repayment when only one of the borrowers in the type 3 group is successful. Hence it will be necessary to increase the interest rate to recoup costs from the instances when both borrowers are successful.
    1. In other words, there may be selection bias. If so, then a difference in income between the treatment group and the control group might be due to this difference in entrepreneurial ability and not because of a difference in the treatment. Hence, I might falsely advise my MFI client that the program was successful whtn it was not.
    2. If I believed that the high interest rates were not due to market failure, but rather due to the monopoly power of the moneylenders, then I would first find out what the source of the monopoly power was. If it was due to a proprietary database, I might introduce legislation to force the money lenders to share this information. If it was due to collusion between the moneylenders and the village chiefs, I might take steps to prevent this collusion.
    3. Credit cooperatives allow for savings without borrowing, and also for more than one person to borrow at a time. However, unlike a bank, all the borrowers and savers are stockholders in the cooperative; hence everybody guarantees all loans. This being the case, the cooperative can obtain funds from outside at cheaper rates, than if the members by themselves were to try and obtain funding for their enterprises.
    4. Competition might not be a good thing in microfinance, because there is a freerider problem. If one microlender takes the trouble of setting up groups, another microlender might step in and use the existing groups to provide loans cheaper. This would undercut the incentive for any microlender to go to the expense of setting up groups. Furthermore, since the loans have no collateral, they work by denying non-payers access to future loans. However, if there is competition, the original lender might not be able to keep defaulters from obtaining loans from competitors. If there are non-profit MFIs that perform the initial work, the resulting free-rider problem could lead to an oversupply of funds since there would be many other for-profit MFIs that try to recoup the benefits of the work already done. This could serve to reduce interest rates to low levels which could lead individuals to take on excessive loans.
    5. The advantage is that bad weather is more easily and more cheaply observable than bad harvests; to observe the latter, it would be necessary to visit each farmer's fields separately. Hence insuring bad weather would work well, assuming that there is not a lot of variability across fields and farmers in a given area; else the basis risk would be too much and farmers might not get effective insurance. Furthermore, there is no issue of moral hazard in insuring bad weather, whereas there is with insuring bad harvests.
    6. Increased income from microloans could make it easier for families to withdraw women from outside activity and thus entrench gender roles.
    1. Friends and neighbors help each other; this works as social insurance. On the other hand, this makes it difficult to engage in efficient saving because neighbors and relatives might come by at any time and request a loan from a nest egg that the person has been building up for a productive purpose, such as to buy a machine. In addition, in many cultures, husbands can forcibly expropriate the savings of wives.
    2. An auction ROSCA allows the person with the most productive investment projects to outbid competitors ensuring that the funds are channeled to their most productive use. Furthermore, there is no intermediary needed.
    3. Health insurance is more difficult to provide than funeral insurance because of several reasons. Death is difficult to dissemble; a person might claim to be sick and to have spent money even when s/he is not sick. More importantly, there may be moral hazard; the person might not engage in healthy behaviors because of the health insurance, thus increasing the cost of providing health insurance.
    4. Moneylenders often do not charge extra if borrowers take longer to repay their loan. Hence the effective rate might be less than the stated rate. Furthermore, moneylenders' interest computations don't involve compounding.

Final Exam


  • Quizzes substituted for Final Exam

Solutions to Final Exam

See quizzes.






Go to Microfinance Home Page