Click here for search results


Site Tools

Using Surveys to Measure Households' Access to and Use of Financial Services

March 2008, Bob Cull (DECRG) and Patrick Honohan (Trinity College, Dublin)

The microfinance revolution has opened-up new opportunities to households across the world.  Most attention has focused on the extension of small loans, often to groups formed by microfinance lenders.  But microfinance has been diversifying its products to include savings media and different forms of micro-insurance as well as payments and other non-financial services.  And it’s not only specialized microfinance entities that have been expanding the provision of financial services to the poor and to micro-entrepreneurs.  For one thing, mainstream banks have become engaged, encouraged by new technologies that can lower the transactions costs of service provision.  Conversely some of the larger microfinance institutions have moved into mainstream banking.  And other classic providers of small-scale financial services, such as credit unions, and postal and other savings banks, remain very important players in many countries.  In addition, mobile phone companies have begun to make inroads into the retail payments business and may go further.

Prompted by the increased interest in this field expressed by policymakers and market participants, and which emerged especially during the recent International Year of Microfinance, the World Bank has taken a leading role in the drive for better information  on household access to finance.  Information from financial service providers and national experts can help in this regard, and progress has been made on these fronts, both with new surveys and with new ways of processing and combining information from different sources (e.g. Beck et al., 2007, Honohan, 2008).  But with the diversity of providers, and the ever-widening range of services, it has become increasingly difficult to rely on industry sources and expert surveys to form a picture of the quality and reach of financial services to low and middle-income households in developing countries. 

Indeed, to get a picture of just how households are dealing with their need for financial services, there is really no substitute for surveying household members directly.  Some specialized surveys have been conducted in the past, and some financial service questions have also been included in a handful of general household surveys in developing countries including some of the LSMS (Living Standards Measurement Surveys program).  The consensus of specialists has been that much more information is needed and that some degree of harmonization across different surveys is highly desirable if a coherent picture is to be built. 

The World Bank, in discussion with funders and survey specialists, has embarked on a wave of household surveys focused on financial services access.  This wave is using a new questionnaire (survey instrument) designed to elicit the relevant information in a comprehensive manner.  The questionnaire is being adapted to local conditions, but will retain sufficient commonality to allow cross-country comparisons to be made with confidence.

As these surveys roll out, we will continue to learn about how best to conduct them. For example, one surprisingly important issue centers around the unit of response.  Preliminary results from randomized experiments in Ghana indicate that 26% of households use banking services when the head of household responds to a survey as opposed to only 10% when a randomly selected adult non-head responds. Based on a full enumeration (i.e., summing survey responses across adults within a household), 26% of households use banking services. Since full enumerations tend to be more costly than surveys of a single respondent, it is encouraging that the responses from the head of household produce nearly identical rates of access. Ghana is but a single country, and so we will extend these experiments to additional countries to see whether similar results obtain, for example in Timor-Leste in late 2008.

The contextual cues contained in survey questions also appear to affect reported use of financial services. The Ghana findings indicate that measured access is higher for some services when questions are asked about specific products (ATM cards, credit cards, life insurance, disability insurance, etc.) than about dealings with types of institutions (banks, credit unions, insurance companies, etc.). The disparity is larger for more complex services – 18% of Ghanaian respondents report personal use of insurance when asked about specific products while only 6% do so when asked about their dealings with providers. By contrast, 14% report that they have a bank account when asked about use of products, just slightly higher than the 13% who were asked about their dealings with banks.

Potential impediments to households’ use of financial services include a limited awareness of their availability and a lack of familiarity with how such services work. We are therefore also conducting a field experiment in Indonesia to explore the impact of financial literacy on households’ willingness to open bank accounts. In the Philippines, we are extending that line of research to examine how the relative financial literacy of wives and their husbands affects the efficiency of investment, as well as how the returns to investment are distributed within the household.

Table 1

Timor Leste

In the field late 2008
Currently in the field
In the field late 2008
In the field 2008






These are only a selection of the technical issues that are being tackled in what is hoped will become an extensive effort.  To date, including the Ghana survey which has already been conducted, nine countries are either under way or planned so far in the Bank’s new wave of surveys (Table 1).  These will complement the Finscope surveys being carried out with a somewhat different methodology.

A new webpage has been created which sets out the issues, and provides links to the surveys and other relevant material.  It will be updated as the findings of new surveys become available.

Availability of this and other survey-based data on financial access in developing countries and the response of individuals to availability and price of financial services is leading to an explosion of data-based research.  The World Bank is planning a conference for Spring 2009 to enable a wide-ranging discussion of research findings.  Advance planning of the program is under way and the organizers invite paper submissions by November 1, 2008 (send them to Mani Jandu,

Beck, Thorsten, Aslı Demirgűç-Kunt and Maria Soledad Martinez Peria. 2007. “Reaching Out: Access to and Use of Financial Services across Countries.”  Journal of Financial Economics 85(1): 234-66.

Honohan, Patrick. 2008. “Household Financial Assets in the Process of Development” in James B. Davies, ed., Personal Wealth from a Global Perspective UNU-WIDER Studies in Development Economics (Oxford University Press) forthcoming.


Permanent URL for this page: