Document Sample
					CASE STUDY: 1st view oct 01

  (carried out an Exploratory Study based on Case Studies of Clients of
  six Micro-finance Programmes in Kenya, entitled: "Financial Services
                           for the Poor" (FSP))

Contact Details:

Charles Marwa
Research Assistant
P O Box 72951
Commercial Street
Nairobi, Kenya

Tel: +254-2-530536/532993/532994                      Fax: 530537


This note is derived from documents provided by Research International -
Kenya, and one interview in July 2001 with Charles Marwa, RI's Research
Assistant who carried out the exploratory study for DFID.


This exploratory study was commissioned by DFID (East Africa) in late 1999.
It was based on structured interviews with 15 clients from each of 6 Kenyan
micro-finance programmes. The study was commissioned to obtain a deeper
understanding of some of the dynamics inherent in the use of financial
services, and their impact on businesses and households. The main source
of data was a questionnaire that allowed both structured and semi-structured
responses. Attribution of impact relied upon qualitative interpretation of
evidence rather than statistical inference: it is clearly stated that the research
is intended to be exploratory and indicative, rather than conclusive and

The attachments, supplied by Research International - Kenya, include a copy
of the questionnaire and the instructions to the interviewer. A sample of a
completed case study is also attached.

This note summarises the lessons learnt from the methodology used in the
study, rather than the specific findings of the study.
CASE STUDY: 1st view oct 01


The British Aid to Small Enterprise (BASE) programme in Kenya has, since its
inception, been strongly involved in the development of the micro-finance
sector. In October 1997, following the review of earlier activities, a new five-
year strategy for the programme was developed. One of the five outputs
planned for the BASE Kenya programme over the five years 1997-2002 was
“increased capacity of private sector providers to deliver financial services to
micro-enterprises, particularly those owned by women and those in the rural
areas, on a sustainable and replicable basis.” In support of this output,
implementation of existing micro-finance projects continued and support was
extended to further micro-finance activities. Consistent with both simplifying
management arrangements and recognizing the need for a broader sectoral
focus, all micro-finance support activities under BASE were incorporated into
a new, larger umbrella project: Financial Services for the Poor (FSP)

A number of BASE micro-finance projects, which were eventually subsumed
within the FSP umbrella project, were time-tabled for mid-term output to
purpose review in 1999. In order to both increase the efficiency of the review
process itself and benefit from cross-project and sectoral learning a combined
review of these projects was undertaken. Assessment of the impact of the
FSP projects formed a key element of review.

The overall purpose of the FSP review was to assess the progress of each of
the DfID projects against the stated purpose and offer guidance for the
implementation of the remaining elements of each project. In combining these
project reviews, the aim was to draw lessons where appropriate from the
comparative performance of each project and from examining the BASE
micro-finance programme as a whole. This could be used both as a basis for
assessing DfID‟s broader involvement in the development of the micro-finance
sector in Kenya and to guide future work. Clearly in drawing comparisons
between projects it was essential to take account of the significant differences
between them.

A Research, Monitoring and Evaluation Project (REME) was established in
1997 to assess the impact of BASE at both programme and individual project
level. Although REME produced some outputs relating to projects falling
under the FSP umbrella, as there was insufficient data at that time to attempt
an assessment of the impact on target beneficiaries of these FSP projects.
Furthermore, it was felt that there was insufficient understanding of the
mechanisms by which FSP services make and impact, to be able to design a
successful impact study.

As such, this study was commissioned to obtain a deeper understanding of
some of the dynamics inherent in use of financial services, and their impact
on businesses and households. This report is a compendium of individual
case studies. Each case study includes a narrative giving an overview of the
story of the respondent and their use of services, an analysis against
hypothesis, and key figures. The nature of the study is, however, to provide
CASE STUDY: 1st view oct 01

the „raw‟ information which can be used in further analysis and which can
perhaps be expanded or followed up on as part of a longitudinal study.


The aims of the exploratory study were to: -

 Understand how different types of micro-finance services affect individual
  clients, their businesses and households.

 Provide information on impact and reaction to services of direct use to the
  Micro-Finance Institutions (MFIs) themselves.

In terms of output the objective was to provide a number of in-depth case
studies, which will form a platform for further analysis, expansion and follow-
up over time.

3     METHOD

3.1   Background on method

It is well established that the impact of access to micro-financial services
cannot readily be understood by examining either the client or the business
(for which a loan may ostensibly have been granted) in isolation. Rather
clients are a part of a wider household and the business may be one of a
number of income generating activities in which the household is engaged.
The impact of access to financial services must be understood in terms of the
effect on the wider household and livelihoods.

In this study, attribution of impact relied upon qualitative interpretation of
evidence, rather than statistical inference. Attribution of impact depends on
understanding, from the client perspective, how each loan or use of a savings
facility has impacted on aspects of their livelihoods. This research is
essentially exploratory and is indicative of how various financial services can
impact on livelihoods rather than conclusive regarding the specific aggregate
impact of the various projects.

3.2   Basic Approach

Because of the complexity of the situation of the business and household, a
qualitative, in-depth method of understanding was considered most
appropriate. The study comprised in-depth interviews with selected clients
using a research guideline incorporating both structured and unstructured
CASE STUDY: 1st view oct 01

3.3      Sample selection

Six MFIs were selected for the study. These were: K REP Bank

     Faulu
     WEDCO
     KWFT: Kenyan Women's Finance Trust
     Co-operative Bank (MSE unit)
     K-REP NGO FSA: (Financial Services Association)

All operate from Nairobi, except K-REP FSA and CARE-WEDCO whose
operations are based in the remote parts of Taita-Taveta, Machakos,
Marsabit, and Bomet Districts; and the Nyanza and Western regions of Kenya

Between December 1997 and February 1998, the Institute of Development
Studies (IDS), University of Nairobi, conducted a previous impact study, called
REME. Under this REME study, four MFIs‟ clients had been included; K-REP
Bank, KWFT, Faulu, and WEDCO. The respondents interviewed within this
FSP Exploratory Study, for these MFIs, were selected from the 300 who had
been previously interviewed. The idea of selecting clients covered under
REME was to allow for the possibility to assess and compare their perception
in-terms of impact on their livelihood as a result of their interactions with the
MFIs and how the loans were actually utilized over time.

Since the K-REP FSA, and the Co-operative Bank had not taken part in the
earlier REME study, fresh sampling had to be conducted to select their clients
for the study. In this case, respondents were randomly selected from the
current client list.

Follow up of REME respondents was problematic. It was very difficult to track
down these clients after such a long time. Some were completely untraceable,
having changed their businesses and careers, migrated or even died. There
were some who had married, and changed their names from the ones
appearing on the list.

3.4      Sample Structure

                Coast    Mt Kenya      Nairobi    Machakos    Western/Nyanza        Total
FAULU             -          6           9           -               -               15
K-REP             -          -          15           -               -               15
WEDCO             -          -           -           -              16               15
KWFT             4           5           6           -               -               15
K-REP            8           -           -           7               -               15
Co-op bank        -         15           -            -              -               15
CASE STUDY: 1st view oct 01

3.5      Field team

To ensure that quality work was achieved, Research International (RI)
selected a team of interviewers who were considered to be both
knowledgeable and experienced enough, to be able to understand the job
requirements. Most of the team members were University graduates with
backgrounds as relevant as possible to the research subjects. Most were
already experienced in field interviewing, and were selected from our Nairobi
or Kisumu teams.

3.6      Training

Training of the interviewers took place within the premises in Nairobi and
comprised four parts:

     In-depth interviewing skills for one day (trained by one of RI's Senior
      Qualitative Research Officers)
     Background on the area of Financial Services provision and impact,
      covering the sustainable livelihood framework (trained by the officer in
      charge of the project at DfID) – one day

     Application of the questionnaire – two days

     Following piloting, the interviewers underwent a further training session on
      the new questionnaire.

The idea was to ensure that the team understood possible ways in which the
financial services could impact on the poverty and vulnerability of clients and
their households, to ensure in-depth probing and follow-up in relevant areas.
They were also trained extensively on how to approach respondents and build
up rapport that would lead to openness in response.
CASE STUDY: 1st view oct 01


After the training, the team was sent to the field for pre-test purposes. During
this first trial, interviewers were paired together; one to be the facilitator and
the other the note taker. From the experiences of this trial, the interview
guideline was subsequently revised and strengthened. James Copestake, a
financial sector consultant for DfID, joined the team for a few days and further
worked towards improving the research tool. The teams wrote out their case
studies and these were later sent to DfID for input and comment.


Nairobi, being the headquarters for most of the MFIs, had the highest
concentration of respondents who were scattered in different parts of the city
and conducted a variety of businesses. Thus the research started in Nairobi
and lasted about eight days. Nyanza and Western regions were covered by
the Kisumu based FSP team. From Nairobi, one half of the Nairobi based
team went to the Coast whilst the other half went to cover the FSA in
Machakos. Mt. Kenya region was the last to be reached.

FSP field activities started on the 1st of December 1999 and were concluded
on the 14th January 2000.

The major problems encountered in the field were too long periods of time
taken to locate some clients and delays occasioned by the rescheduling of the
appointed interviews because the clients were very busy in their businesses.
Other problems included clients who did not divulge correct information about
their wealth status or other details. Some MFIs listings their projects as being
in Nairobi, covered districts further away, including Machakos, Kitui and Thika.
Tracking of sampled clients also had to be extended to such places. This cost
considerable time.


The final interview guide / questionnaire is attached. This was designed in
such a way as to allow respondents to express issues they felt important, and
to collect basic facts and figures on households and business. As such it
comprised a series of unstructured and structured questions. Interviewers
were encouraged to probe throughout the questionnaire, even where an issue
arose during a structured question, and to insert this additional information
into the open sections.
CASE STUDY: 1st view oct 01


7.1   Case Study Structure

The process of analysis of the case studies involved summarizing the life
story of the respondent as told by them. This was constructed into a
summarized, not verbatim, narrative. Following completion of the narrative, a
separate analysis of each case was completed against the hypotheses
agreed with DfID. The third part of each case comprises a table of key data.

Information from the structured section of the questionnaire has been entered
into an excel spreadsheet, which can be used for further analysis (see

7.2   Analysis against Hypotheses

RI agreed with DfID a set of hypotheses about the impact of financial
services, against which each case study was `tested‟. Looking at each
interview, comparing qualitative and quantitative sections, observations and
key data and then making a conclusion against each hypothesis was the
methodology employed. This is in bullet form at the end of each narrative
case history.

The main hypotheses that this research will seek to explore are as follows:

      i)        Impact on the poor

             Micro-finance services have been taken up by a highly diverse
             range of clients including many who are poor and many women.

      ii)       Business Improvement

             Loans have helped some clients to diversify their business
             activities, to increase the capital invested in their businesses, to
             increase paid business employment, to increase business
             profitability and generally to decrease the vulnerability of the
             business to shocks.

      iii)      Household livelihood Improvement

             The income of some clients has gone up enabling them to increase
             expenditure on food, health services, education, shelter and
             household assets and to cope with livelihood shocks.

      iv)       Reduced household vulnerability

             The vulnerability of clients and clients‟ households has been
             reduced by the increased security of incomes from business
CASE STUDY: 1st view oct 01

              activities or the creation of accessible financial assets to protect
              against risks and to cope with livelihood shocks.

       v)        Increase in paid employment

              Loans have facilitated the expansion of some micro enterprises
              (which may be owned by non-poor clients) increasing the paid
              employment for the poor people.

       vi)       Enhancing women’s status and livelihood

              Access to micro finance services has helped to strengthen the
              status and livelihood of women

       vii)      Improved impact on the poor through developing the range
              and terms of financial services available.

              Impacts, particularly on the poor and on women can be improved
              through further developments of the range and terms of financial
              services on offer.

A sample case study, including narrative and analysis against the above
hypotheses is attached.


When they came to be analysed, responses (sometimes contradictory, other
times non-existent for some questions) revealed several problems with the
research methodology, both the in the format of the questionnaire, and the
way interviews were conducted. These included:

   it was possible for interviewers to 'lead' the client towards certain answers
   some interviewers did not know enough about business to draw out
    answers, and calculate business income
   there was not enough cross-checking of answers
   sources of income were confusing, sometimes contradictory, and often not
   there were problems in classifying the type of business; also the number
    and pay of employees (for example, due to misunderstanding or varied
    interpretations of 'paid' and 'employee')
   one interviewer noted that one woman was reluctant to answer because
    her husband was present
   in completion of the 'before' financial sections of the questionnaire there
    was strong reliance on memory
   some had received loans from elsewhere so there was confusion about
    which loan had achieved the impact
CASE STUDY: 1st view oct 01

At the start of the analysis the Research Assistant had to try to deal with the
anomalies he found by checking and cross-checking with the client and the
interviewer; he admittted, however, that inevitably some assumptions had to
be made from the responses, and that writing up the case studies was
sometimes difficult due to the contradictions in the data.


The Research Assistant commented that at first some of the institutions had
'glanced over' the 'improvements in services' suggested by some clients, but
over the duration of the analysis period the institutions came to realise it was
critical to take the 'improvements' seriously as this was part of the learning
process. He had found one of the six institutions was wary of the study and
its outcomes at the start of the research, but the other five all co-operated


If further research is undertaken, efforts will be made to overcome the
problems found in the 1999/2000 study, but since these were not found to
have had a major outcome on the overall conclusions of the study, the
methodology will remain basically the same. The following methods are
therefore proposed for the next Study which will take place in 2003 at the end
of BASE's 5-year programme:

    some of the case studies will be selected for follow up, by type of business
     and geographical region: it is recognised that it will be hard to follow up all
     clients used in the first study as some will have moved. Most did seem to
     be established businesses, however, and it should therefore be possible to
     find sufficient to enable a meaningful follow-up study

    will use a similar format of questionnaire, except the wording will be
     clarified to minimise misunderstanding / variable interpretation of some