Database and Development Indicators

Philippines
Improving Poverty Monitoring System

To gather social, demographic, health, labor force and employment statistics, the Philippines National Statistics Office (NSO) conducts a system of household surveys that is based on a master sample taken from the most recent census of population and housing. These surveys have many other critical uses including providing inputs for economic and social accounting as well as poverty monitoring and analysis. The two surveys in this system of surveys that are most concerned with poverty monitoring are the Family Income and Expenditure Survey (FIES) and the Annual Poverty Indicator Survey (APIS). FIES collects data on family income and living expenses and related information affecting income and expenditure levels and patterns. It is the basis of poverty estimates and of the weights for the consumer price index. The two more recent rounds of FIES were designed to provide reliable estimates at the provincial-urban/rural level. On the other hand, APIS was initiated in 1998 to provide more frequent assessment of poverty status, specifically during the non-FIES years. It collects information on the socio-economic profile of households and other information relating to their living conditions - economic characteristics, access to electricity, health, education, housing, available water and sanitation facilities and availability of credit to finance family business or enterprise.

Although both FIES and APIS collect income and expenditure data, the estimation of the incidence, magnitude and depth of poverty was not yet attempted using APIS because of incomparability of some data items and reference periods between the two surveys. Furthermore, while the common components of the master sample were used in both surveys, the results of FIES and APIS were analyzed independently and not as a panel, thereby missing the opportunity to measure the gross change in poverty status of households and the insights that could be gained from investigating the relationships of the gross change of poverty status to other household characteristics.

To support the Government's poverty reduction program, NSO plans to continue conducting APIS in the intervening years of FIES. NSO conducted the 1999 APIS entirely from Government's funding. Annual poverty indicators could therefore, be compiled from these two surveys provided that the problems mentioned above are resolved. However, while monitoring poverty is a government priority, the cost, hence sustainability, of conducting FIES and APIS regularly have not been investigated seriously. Moreover, there may be insignificant value added to measuring poverty indicators and correlates yearly since the annual movement of these indicators may not be detected or may not be statistically significant because changes in these indicators are usually small relative to sampling error. There is a need, therefore, for an in-depth study on how these surveys could be more efficient as well as be cost effective.

The capacity of NSO to design and implement sample survey strategies must also be strengthened. Although many of NSO technical staff have academic background in sample surveys, their experience is limited only to the sampling design and operations of NSO surveys. Aside from the usual tabulation of survey results, there has been little research and analysis done on the sampling errors of completed surveys with the purpose of enhancing them. NSO has tended to rely on experts from international donor agencies to develop and improve the master sample design for the system of household surveys. Because these experts were only available intermittently, skills transfer to NSO staff had been slow or inadequate that further improvement to the master sample design still needs to be done with external advice.

The goal of this TA is to improve NSO's poverty monitoring surveys. The TA aims to assist the NSO in achieving the following objectives:

1. Link FIES and APIS for more in-depth poverty analysis and research;
2. Improve the efficiency and cost-effectiveness of FIES and APIS by developing better survey strategies (questionnaire, sampling and field operations); and
3. Establish a research and development unit at NSO that will be responsible for sustained enhancement of survey sampling and operations.

Status of the project: being processed; not yet approved