The author recently completed an internship in Fonkoze’s Ma

the project discussed in this paper, to the Institut Ha?tien de Statistique et d’Informatique for providing access to national survey data, and to Plan International for funding the collection of data on Fonkoze’s social performance. Thanks are also due

Rob Fuller, 26 September 2006

The author recently completed an internship in Fonkoze’s Market Research and Social

Performance Monitoring Department.

Fonkoze is grateful to Grameen Foundation USA for providing technical assistance with

the project discussed in this paper, to the Institut Haïtien de Statistique et d’Informatique

for providing access to national survey data, and to Plan International for funding the

collection of data on Fonkoze’s social performance. Thanks are also due to Mark

Schreiner for providing detailed comments and assistance with this paper.

Fonkoze, a microfinance institution in Haiti, has used a poverty “scorecard” to

benchmark existing client data to the $1-a-day extreme poverty line. By referring to data

from a national survey of household income, 12 variables already routinely collected by

Fonkoze were found to be suitable as indicators of whether a household is above or

below the extreme poverty line. A scorecard was constructed, which allocates a number

of points to each household depending on their responses to those 12 questions: this total

score can then be translated into a probability that each household lies above or below the

poverty line. This scorecard has already been used to analyse poverty rates of clients

across several Fonkoze branches and credit cycles.

the project discussed in this paper, to the Institut Ha?tien de Statistique et d’Informatique for providing access to national survey data, and to Plan International for funding the collection of data on Fonkoze’s social performance. Thanks are also due

Fonkoze is Haiti’s largest microfinance institution, serving clients throughout

rural areas of the country. Since its creation in 1994, Fonkoze has grown to a network of

29 branches, serving over 40,000 female credit clients through its solidarity loan

program.

Fonkoze has long had a commitment to tracking the impact of its work on the

lives of its clients. Its primary method for collecting information on the wellbeing of its

clients is the Kat Evalyasyon (“evaluation card” in Creole), which is used to assess all

clients on entry to the credit program, and then at intervals as they progress through the

program. The tool has been adapted over several years, and now consists of questions on:

Materials used in the construction of the family’s house

Type of bed and toilet/latrine used by the household

Ownership and use of land

Ownership of animals

Ownership of household goods

Husband’s work

Whether the household receives transfers from overseas

Forms of transport used

Client’s level of literacy

Whether children in the household attend school

How often the household eats meat.

Fonkoze uses this information to build a broad picture of the characteristics and

living conditions of its client base. What it could not do, however, was to provide a

the project discussed in this paper, to the Institut Ha?tien de Statistique et d’Informatique for providing access to national survey data, and to Plan International for funding the collection of data on Fonkoze’s social performance. Thanks are also due

quantitative measure of clients’ poverty. Following the recent US legislation on funding

for microfinance, it has become of interest to be able to produce quantitative estimates of

institutions’ poverty outreach – that is, what proportion of institutions’ clients are living

below a certain poverty line. From a management point of view, it is also of interest to be

able to compare poverty outreach within an institution, between branches and inpidual

credit agents, and to study how the proportions of clients in poverty changes over time.

How is it possible for microfinance institutions to determine how many of their

clients are living below the poverty line? Measuring poverty status directly is a

complicated, time-consuming and expensive process, which involves asking families

questions on all their perse sources of income (or, alternatively, on their perse

household expenditures) with as great an accuracy as possible, in order to calculate

whether each household’s income or consumption is above or below the poverty line.

Fonkoze instead followed the approach developed by Schreiner et al. (2005) in

developing a poverty “scorecard”, consisting of a list of simple questions, each of which

is an indicator of household poverty. For example, a poverty scorecard may include

questions on the condition of the house itself, since poor quality houses tend to be

associated with poorer families. By using data from an existing national household

survey, each question can be allocated a certain number of points, and the scorecard

calibrated so that the total number of points for a household can be translated into the

probability that the household is living below the poverty line. The advantage of this

method for the microfinance institution is clear: the institution only needs to ask each

client a few straightforward questions, rather than doing a complete survey of their

household income or expenditure. For Fonkoze, there was an even greater benefit to be

the project discussed in this paper, to the Institut Ha?tien de Statistique et d’Informatique for providing access to national survey data, and to Plan International for funding the co …… 此处隐藏：24043字，全部文档内容请下载后查看。喜欢就下载吧 ……