In This Issue

Workshop Session: Depth of Outreach: The New U.S. Law Requiring Cost-Effective Measurement Tools

Workshop Session: Ownership and Governance in Microfinance

Register now for the Global Microcredit Summit 2006

Archived Issues

Vol 3 Iss 2 October '05
Vol 3 Iss 1 April '05
Vol 2 Iss 4 Dec. '04
Vol 2 Iss 3 August '04
Vol 2 Iss 2 July '04
Vol 2 Iss 1 June '04
Vol 1 Iss 6 Jan. '04
Vol 1 Iss 5 Nov. '03
Vol 1 Iss 4 Sep. '03
Vol 1 Iss 3 July '03
Vol 1 Iss 2 May '03
Vol 1 Iss 1 March '03

» Current Issue

E-News Information

Reprinting Permissions

Subscribe to Microcredit Summit E-News

Workshop Session: Depth of Outreach: The New U.S. Law Requiring Cost-Effective Measurement Tools

Remarks by Stacey Young

As Joanne already explained with regard to the history of this law the working definition of very poor that's enshrined in the law is of the very poor, those living on less than a dollar a day or in the bottom fifty percent below their country's national poverty line. And what's at issue is USAID microenterprise funding and it's important to keep in mind that the way that we work in microenterprise development involves microfinance but also what we call business development services and enabling environment activities which are aimed at policies and regulatory framework issues. One of the key points of this mandate is that it involves absolute measures of poverty as opposed to relative. Many of the existing tools for poverty measurement deal with ranking people in relation to other people in their communities. These are relative measures of poverty. This law mandates that we measure absolute poverty. The law also indicates that the tools should be both low cost and accurate and that they should be field tested before certification so that we can use the field testing process to determine that they are low cost and accurate. And they must be applicable in a wide variety of circumstances both because we work at microenterprise development in a lot of different ways you know through financial services as well as BDS, an enabling environment, but also because we work in all four major regions of the developing world.

So the tools that we are developing are comprised not only of the indicators that we've been discussing a lot during the course of this conference, but also guidance for implementation of the tools and also guidance for data entry and analysis. So what institutions that receive USAID funding will be getting from us is not just a list to measure this, that, and the other indicator, but guidance on how you do it and then here's what you can do with the results. We will be developing a training manual for users, but we haven't yet determined whether we'll have resources available for actual in person training for people who need to implement the tools.

So the challenge embodied in this mandate is to identify who the very poor are. This is difficult because very poor households have non-standard sources of income and also because it's very difficult to get people to report directly on their income and in fact, in many cases they may not even know what their income is because it may take the form of various assets as well as cash.

Expenditure surveys that have been developed to measure income or to measure poverty or assets are expensive and they take a lot of time and hence the need to develop shortcut tools. The players in all of this include IRIS, which is the Center for Institutional Reform in the Informal Sector. It's an institute located at the University of Maryland and we're working with them on developing the poverty measurement tools. They have a lot of experience in this kind of work. So they're participating and we also have established an advisory panel comprised of academics and practitioners with experience in this area. So, they are also participating. These two groups of participants supply the methodological experience and rigor needed to develop tools that will be precise according to the terms of the law.

Also we get practical insight through our consultation with the advisory panel and also members of the SEEP network's poverty assessment working group. The SEEP network is a small enterprise education and promotion network in the United States. I do want to mention also that both Anton and Chris are members of the advisory panel. And then, of course, USAID plays a role in this by supplying funding management and coordinating the consultative process as well as looking at what the next steps are in terms of implementation and beyond.

And so the process that we have followed over the past year is as follows: We've laid the groundwork first for the consultation. The law mandates, and also common sense dictates, that these tools shouldn't be developed in a vacuum but rather in close consultation with the people who actually are going to be tasked with implementing them. And so the consultative process was key and that was something that we had to put in place at the very beginning, so that involved constituting the advisory panel and working with SEEP in order to get the Poverty assessment Working Group (POWG), a subgroup of SEEP members on board with this activity and contributing their insights. And that consultative process has continued and will continue throughout the entire period of tools development. We then solicited tools that practitioners are already using and from those tools we basically looked at all the different indicators that practitioners are using currently to measure the poverty steps of their clients and we just kicked out the duplicates. So we ended up with a long list of potential indicators. Then we worked again in consultation with the advisory panel and the POWG members, to establish a set of criteria which would form the basis for certification of tools. In other words we need to have everybody agreeing from the outset as to what constitutes a low cost, accurate poverty measurement tool. From that we developed a composite survey and this [used] the indicators compiled from the tools that practitioners submitted as well as indicators that appear prominently in the literature on poverty measurement. So, these are indicators that come out of experience with poverty measurement.

And then the first step for testing the tools was to test them in terms of accuracy, keeping in mind that the tools have to be both low cost and accurate because what we're looking for is remember proxy indicators; indicators that we look at that take the place of information that we can't get or that we can't get in a low cost manner. We need some proxy indicators that will tell us in a more efficient manner, do these people fall above or below that line that we're drawing at a dollar a day or the bottom fifty percent below the national poverty line. For the purposes of the accuracy tests we used local survey firms and the reason for this was that we wanted to isolate the variable here that is the indicator. We needed to use survey firms that were experienced in conducting surveys so that we didn't end up testing the ability of the tester but rather testing the accuracy of the tool. And so the survey firms came from within the four countries in which we conducted the accuracy tests. We used one country in each of the four regions that USAID operates, so Bangladesh for Asia, Peru for Latin America and the Caribbean, Uganda for Africa, and Kazakhstan for Europe and Eurasia.

The next step is practicality tests, determining whether the tools are actually practical for use in the field. We don't want to end up with a tool that nobody can actually use because it requires them to sit down for three hours with each client or because it requires them to invest a lot in staff members when that would burden their financial sustainability.

We have some preliminary sets of indicators and we've integrated them into tools, again keeping in mind that the tools constitute not only the indicators but also guidelines for implementation and for reporting on results. We're in the process right now of selecting practitioners from among those who have applied to conduct the practicality tests. The tests in this case will be conducted by practitioners themselves because we need to see how they work when they're not, conducted by survey firms, not by professional surveyors but by practitioners themselves. And they will report to us on cost, ease of adaptation and applicability. And we anticipate that about ten to fifteen tests will be run in 2005. When we get the results we will certify the tools based on both the accuracy and the practicality test results. And from there we'll go forward with implementation.

We have some questions around implementation that remain. One question is what kind of technical assistance will we be able to provide to our implementing partners for implementation of these tools? We will be developing a manual. Will that be sufficient? We don't know. Will we have resources to conduct training, for example, or training of trainers? We don't know. So that's a question that remains outstanding. We also don't know what, if any, steps we can or will take to validate the accuracy of the tests when they're actually being implemented by our implementing partners. As I mentioned, because we were working to establish the accuracy of the tests themselves we did the accuracy tests with professional survey firms. Will we lose some of the accuracy when they're implemented by partners? Quite possible. Will we have the resources to go back and validate the accuracy of the tools when they're being implemented on the ground? We don't know. There are other issues as well. How do we identify the bottom fifty percent below the national poverty line? Again keeping in mind that we're not talking about people living at a dollar amount that's half of the poverty line. … We're talking about a population distribution issue. The bottom fifty percent of all the people who live below the poverty line who likely are not distributed evenly throughout that. And so how to determine where the cutoff point is for the bottom fifty percent is a thorny issue and something that would have to be done on a country by country basis and on the basis of other data that, for example, the government had already collected and that we could verify as being accurate.

Some questions that we can discuss today or, or any other time, questions that have come up in the course of public conversations about this process are: What are the advantages and disadvantages of a legal requirement that USAID recipients use a particular method to measure their client's poverty? Is a legal mandate based on specific metrics necessary given the importance of reaching the very poor? Is it likely to yield the most impact? Another question, what are the, what would be the costs and benefits of investing in a flexible approach that allows for developing a variety of poverty regime methods, testing them, and sharing lessons learned about their utility and impact as part of a broader agenda of creating incentive and appropriate products and services to reach the very poor? And a third question, does a legal mandate to measure clients' poverty status constitute an incentive to reach the very poor?

And finally a few caveats. Poverty measurement in itself does not equal screening loan applicants to target the very poor. This is an important point. Some people have looked upon this as a way to very efficiently screen out applicants, loan applicants who fall above that cut off point. But we are cautioning against this because one runs the risk of course of skewing applicant's answers because they perceive rightly or wrongly that an answer in a particular direction will increase their chances of getting a loan. And so what we're saying is that using these tools as part of a loan application process is probably not a good idea in most cases and, in any case, needs to be done very carefully and in conjunction with a lot of other indicators.

The second point, poverty measurement cannot substitute for designing and delivering products and services tailored to the needs of the very poor. We still need to do work on this. Do we effectively reach the very poor now? I think we would all agree no and that's why we have agreed that we need to put more attention in that direction. But does simply measuring the poverty status of our clients yield products and services that address their needs? No.

Finally poverty measurement does not yield the means of scaling up financial services for the very poor, the poor, and the vulnerable not poor that's necessary to achieve poverty reduction. We still have a long way to go in terms of ensuring that all of the people in these groups have secure access to financial services.

Read remarks by Chris Dunford