Summary Information
TF097641 / KCP II - Can Microfinance Foster Entrepreneurship in Poor Communities
KCP Window / Investment Climate & Trade and Integration
Country / Egypt / TTL Name / David McKenzie
Project Period / 06/15/2009 –03/31/2012 / Approving Manager / Asli Demirguc-Kunt
Grant Amount / $ 75,000.00 / Disbursement / $ 73,784.06

Research Team (with affiliations and country of origin for non-Bank members)

The core team was led by David McKenzie (DECFP) and Tara Vishwanath (MNSED) and assisted by Matthew Groh (DECFP).

Research Objectives

  1. Summary of Aims, Objectives and Significant Achievements

Objective: The objective of this KCP grant was to support an impact evaluation of microfinance tied to the expansion of microfinance planned under a World Bank loan project in Egypt - Expanding Access to Finance in Egypt.

Significant achievements including innovations:

  • Despite the massive upheaval in Egypt as a result of the revolution, and changes in the project as a result, an impact evaluation is still underway.
  • Re-designed the impact evaluation to evaluate an innovative micro insurance pilot developed in coordination with an Egyptian NGO
  • Completed baseline surveys for 3000 micro entrepreneurs
  1. Nominated Outputs
  • None to date due to major difficulties and delays in launching the impact evaluation
  1. Major Difficulties[describe any expected or unexpected issues and how the research team dealt with them, and the end result.]
  • This research project was based on an impact evaluation tied to the World Bank loan project: Expanding Access to Finance in Egypt, but unfortunately the original project experienced major, unexpected delays, and then the component that was to be evaluated was put on hold following the revolution. The impact evaluation was contingent on plans for an expansion of microfinance set to begin in September 2010 that were delayed due to factors outside of our control involvingour local government partner, Social Fund for Development (SFD), which had to coordinate with both the post office authority and local NGOs for the planned expansion. This experienced repeated delays due to bureaucratic issues between the three bodies.
  • Egypt’s revolution pushed everything back, disrupting the component that the evaluation was planned around, and political changes led to a high rate of turnover withinour local partners disrupting the project.
  • In April 2011, SFD and a local lending institution told us that they would be launching the microfinance expansion in the Governorate of Menya, so we conducted a baseline there only to find out later that the local lending institution was ineligible for onlending for various reasons, and so this baseline survey could not be used for a follow-up intervention.
  • From April to December 2011, numerous attempts to launch a geographical expansion in a controlled, randomized manner failed. SFD could not find an NGO to which it could on lend through the post offices or in previously untouched areas, and microfinance organizations were reluctant to expand into new areas given the considerable post-revolution uncertainty.
  1. Surprises[describe any outcomes of the research, beneficial or otherwise, that were unexpected at the outset. Further details should be included in the full report.]
  • Coordinating with multiple government organizations requires even more time for an evaluation than anticipated: This microfinance impact evaluation concept was designed in fall of 2009 and set to be implemented in September of 2010, but delays led to more delays and the intervention that was originally set to be evaluated never got off the ground. It is conceivable that it may still happen sometime in the next two to three years, but funding deadlines are not that flexible.
  • Expanding microfinance through the post system because of its widespread infrastructure is not feasible because of its weak infrastructure:This evaluation was based on an expansion of microfinance through Egypt’s postal system, but early on, the main obstacles were the lack of connectivity and lack of computers.
  • There is considerable scope for microfinance product innovation: In response to the problems that this project experienced, we brainstormed as to what evaluations would be possible, and were pleasantly surprised to find out how much enthusiasm local microfinance NGOs had for our planned microinsurance product.
  1. Full Completion Report
  • Background and motivation

The World Bank loan project entitled, Enhancing Access to Finance in Egypt, was intended to expand access to finance to MSEs in Egypt. According to the Project Appraisal Document, the Central Bank of Egypt (CBE) states that over half of credit extended to the private sector goes to less than one fifth of one percent of their clients. On the other hand, CBE estimates that only one percent of total bank credit goes to MSEs.Microfinance is still a very young industry throughout the world and especially in Egypt. There are many poor villages with no formal access to credit. This project was intended to boost the microfinance industry. One hundred fifty million dollars was intended for the microfinance sector, and approximately thirty million dollars was intended for the post office expansion upon which our evaluation was to be conducted.

In particular, this impact evaluation was originally intended to provide valuable knowledge and insights into the impact of expanding microlending to people who had previously not had the opportunities to take out small loans. At the time of design, in 2009, there were only two existing randomized evaluations of microlending. Today, there are at least five (India, Phillipines, Mongolia, Morocco, Bosnia & Herzegovina). While a further study on microfinance expansion would still be of interest, the increasing number of studies which have looked at this question and the difficulty in getting microfinance organizations to expand to new areas in Egypt after the revolution made this less promising.

The aftermath of the Egyptian revolution has been one of considerable uncertainty for households and firms, with a fall in tourism numbers, uncertainty about subsidies potentially getting removed, and considerable political uncertainty. As a result, many firms may be reluctant to borrow or to invest in new capital. With this motivation in mind, we attempted to innovate and design a new financial product – micro-insurance against macroeconomic uncertainty, and worked with the Alexandria Businessmen Association (ABA), a beneficiary of the Enhancing Access to Finance project and the biggest, most efficient microlending organization in Egypt to design this product, launch it to clients, and conduct a randomized evaluation to measure the business and social welfare outcomes of ABA’s clients.

  • Original objectives of the research, along with any changes and their justification; note their consistency with Bank, and KCP objectives

The overarching objective of this research was to conduct a rigorous randomized microfinance evaluation in parallel with the Enhancing Access to Finance in Egypt loan project. The project was intended to expand access to microfinance to poor regions, and to introduce new financial products. Learning about the success of these efforts is consistent with both the objectives of the World Bank loan funding the evaluation, as well as with the KCP’s private sector questions of interest.

As noted above, the failure of the project to expand microfinance to poor areas forced us to change the research design, which is now instead focused on evaluating a new financial product, which is still consistent with the first objective in the PAD to evaluate “the development of alternative financial products which are designed to overcome the low uptake of existing finance.”

  • Assessment of the extent to which the objectives have been met, with explanations of any shortfall.

Due to long delays and political uncertainty previously mentioned, we have not completed the impact evaluation. In November 2011, ABA and the World Bank began discussions about running an impact evaluation alongside ABA’s microinsurance product, but it took 4 months to set up the logistics of everything. Now, we are in the middle of evaluating the microinsurance product that ABA is currently offering its clients. We completed the baseline on April 1st, and we expect to conduct the endline in August 2013.

  • Methodology and data

The revamped project is a randomized experiment involving 3000 microentrepreneurs in Alexandria, Egypt who currently have loansdue to Alexandria Businessmen Association between April 15th and July 30th and who’s current loan value is between 1500 LE and 10000 LE. The 3000 microentrepreneurs are randomly split evenly into two groups: the treatment group that is offered the opportunity to purchase microinsurance and the control group that is not.

Traditionally microfinance clients have at most been offered health and life insurance, and in rural areas, weather index insurance. Yet firms worldwide note that economic and political uncertainty are key constraints to business growth. Our innovative new insurance product insures firms against the risks of sudden inflation, a reduction in sales due to the imposition of a curfew, and economic and political events that would lead to the stock market getting shut down.

We stratified the randomization by branch location and gender, and then created pairwise matches using Mahalnobis distances based on baseline characteristics.

Based on a preliminary market research survey, the mean monthly income of microentrepreneurs in Alexandria is 775 LE, the standard deviation is 250 LE, and take-up of microinsurance should be around 75%. By assigning 1500 people to the treatment group and 1500 people in the control, we have slightly over 90% power to measure a change in 5% of profits, which doesn’t account for the increased power that stratification should provide.

The endline survey is expected to take place in July 2013.

  • Results and outcome: What have we learnt so far?

This impact evaluation is not yet complete. To date, we’ve learned that there appears to be a high demand for microinsurance against macro risks. So far, more than 80% of the treatment group that has had an opportunity to purchase the microinsurance has purchased it, although it is still very early on in the roll-out.

  • Dissemination details (including future plans)

In the fall of 2013, we plan to host multiple seminars with SFD, policy makers and microfinance organizations throughout Egypt to disseminate results.

  • Impact—within and outside the Bank

This impact evaluation is not yet complete. This is an innovative new product, designed to address one of the fundamental constraints firms in many developing countries say they face. If it is successful in changing firm loan and investment behavior, it would therefore be expected to potentially have sizeable impacts and be copied elsewhere.

  • Implications for future research

The project is still ongoing, and so lessons for future research are still to be derived.

  • Capacity building (Have you used local consultants or firms? Have you provided training to client country counterparts?Number of developing country researchers and institutions engaged with KCP research, breakdowns of resources used for capacity building out of total disbursements)

The effort invested in this impact evaluation has built capacity throughout Egypt among a wide array of stakeholders: the government (SFD and Egypt Post), local microcredit NGOS (ABA, BEST, Eve of the Future, and Beni Suef CDA), and a local survey firm (El Zanaty). These stakeholders have been ingrained with an understanding of what and why randomized impact evaluations can be helpful and the basics of how to successfully conduct one. We provided extensive training to El Zanaty on how to implement surveys. Around $7,000 was spent on capacity building these various stakeholders. ABA now has an innovative, urban insurance product

  • Additional resources leveraged

In addition to the KCP fund, resources have been leveraged through DEC, the TFESSD, and 3ie to ensure adequate funding through the endline.

  • Operationalization and development policy implications

There aren’t any development policy implications yet, but this is the first urban microinsurance randomized experiment in the world. Thus, there are lots of potential development policy implications.

Annex 1: Reference list and background papers

Armendáriz and J. Morduch (2005), The Economics of Microfinance, MIT Press,

Cambridge, MA.

Augsburg, R. De Haas, H. Harmgart and C. Meghir (2011), “Microfinance at the margin:

Experimental evidence from Bosnia”, mimeo.

Banerjee, T. Besley and T. Guinnane (1994), “Thy neighbour's keeper: The design of a

credit cooperative with theory and a test”, Quarterly Journal of Economics 109, no. 2:

491-515.

Banerjee, E. Duflo, R. Glennerster and C. Kinnan (2010), “The miracle of microfinance?

Evidence from a randomised evaluation”, BREAD Working Paper No. 278

Banerjee, A. and S. Mullainathan (2010), The shape of temptation: Implications forthe economic lives of the poor, MIT Working Paper.

Bruhn, M. and B. Zia (2011), Stimulating managerial capital in emerging markets:The impact of business and financial literacy for young entrepreneurs, World BankPolicy Research Paper No. 5642, World Bank, Washington DC.

Bryan, Gharad, Ambiguity and Insurance," Nov 2010. Technical Report, mime

De Mel, S., D. McKenzie and C. Woodru(2009), Are women more credit constrained? Experimental evidence on gender and microenterprise returns, AmericanEconomic Journal: Applied Economics, 1-32.

Fafchamps, M., D. McKenzie, S. Quinn, and C. Woodru(2011), When is capitalenough to get female microenterprises growing? Evidence from a randomized experiment in Ghana, World Bank Policy Research Working Paper No. 5706, World Bank,Washington D.C.

Giné, X., P. Jakiela, D. Karlan, and J. Morduch (2010), Microfinance games, American Economic Journal: Applied Economics, 2, 60-95.

Karlan, D. and J. Zinman (2011), Microcredit in theory and practice: Using randomized credit scoring for impact evaluation, Science, 332(June), 1278-1284.

Pitt, M. and S. Khandker (1998), The impact of group-based credit programs on

32poor households in Bangladesh: Does the gender of participants matter?, Journal of

Political Economy, 106(5), 958-98.

Rajan, R.G. and L. Zingales (1998), Financial dependence and growth, AmericanEconomic Review, 88(3), 559-86.

Annex 2: Policy briefs (if any)

The evaluation is still in progress. No policy briefs yet.

1