Official Methodology for the Estimation
of First Semester Provincial Poverty Statistics

This set of procedures corresponds to the official methodology for the estimation of first semester provincial poverty statistics, consistent with the annual official poverty estimation methodology, which was approved through NSCB Resolution No. 9, Series of 2011.

I. Estimation of the First Semester Per Capita Provincial Food Threshold

The estimation of the first semester per capita provincial food threshold starts with the costing of the provincial food bundles. These provincial food bundles were developed based on an indicative nationally-representative food bundle formulated by nutritionists from the Food and Nutrition Research Institute and were subjected to the Test of Revealed Preferences such that the food bundle of a province will be the cheapest in comparison with the bundles of other provinces. It has the following characteristics:

• Nutritionally adequate, that is, it satisfies the 100% Recommended Energy and Nutrient Intakes (RENI) for energy and protein and 80% RENI for vitamins and minerals;

• Food items in the food bundle are locally available and low cost;

• Least cost ¹;

• “Visualizable”; and

• Edible

 

The cost of each provincial food bundle is estimated using the average first semester provincial retail and farmgate prices collected by the Bureau of Agricultural Statistics (BAS) for agricultural commodities and the National Statistics Office (NSO) for non-agricultural food items.  The daily cost of the food bundle is then multiplied by 30.4 days (the average number of days in a month) and 6 months to come up with the First Semester Per Capita Food Threshold. 

II. Estimation of the First Semester Per Capita Provincial Poverty Threshold

The poverty threshold shall still be indirectly estimated as follows:
PT(per capita) =  __   FT________
                        Ave. FE/TBEAnnual

where:

PT(per capita) = First Semester Per Capita Provincial Poverty Threshold
Average FE/TBEAnnual = Average FE/TBE in 2000, 2003, 2006 and 2009
FE = actual food expenditure of families within the +/- ten percentile of the food threshold
TBE = total basic expenditure of families within the +/- ten percentile of the food threshold.  TBE is an aggregate of expenditures on:

• food;

• clothing and footwear;

• fuel, light and water;

• housing maintenance and other minor repairs;

• rental of occupied dwelling units;

• medical care;

• education;

• transportation and communications;

• non-durable furnishing;

• household operations;  and

• personal care and effects

 

The average of the nationally determined FE/TBE ratio from the 2003, 2006 and 2009 FIES shall be used for twelve (12) years starting 2009.  This is equivalent to 0.6983.

III. Estimation of First Semester Incidence and Other Measures of Poverty

Poverty Estimates, which include:

• • Subsistence Incidence (families and population);

  • Poverty Incidence (families and population);

  • Magnitude of Food Poor (families and population);

  • Magnitude of Poor (families and population);

  • Income Gap;

  • Poverty Gap;  and

  • Severity of Poverty

shall be generated and released at the national, regional and provincial levels.

Further, measures of error such as standard error, confidence interval and coefficient of variations shall be released, which will guide the users in utilizing these estimates.

In addition, ranking of provinces shall not be released due to overlapping of confidence intervals.  A clustering of provinces based on the poverty measures, such as poverty incidence and magnitude of poor, shall be done instead.

IV. Computation of Back Estimates

The official poverty statistics for first semester for years 2003, 2006 and 2009 shall be released following the practice among statistical offices of releasing back estimates whenever improvements/changes in the methodology is adopted.

To facilitate the computation of first semester food thresholds for years 2003 and 2006, these will be estimated by deflating the 2009 first semester food thresholds using 2006-based regional Consumer Price Index for Food, consistent with the annual official poverty estimation methodology.  Estimation of the other poverty statistics shall follow the same methodology as stipulated in this document.

V. Estimation of the Subsistence Incidence and the Magnitude of Food Poor

The estimated First Semester Per Capita Food Threshold for each province, urban-rural, is then compared with the first semester per capita income of the families in the FIES.  Families are then classified as food/subsistence poor if the first semester per capita income is below the first semester per capita food threshold.  Subsistence incidences among families/individuals are then computed by counting the number of families/individuals tagged as food poor divided by the total number of families/population, that is:

where F = total number of families/individuals with first semester per capita income below the first semester per capita food threshold
n = total number of families/individuals   

VI. Estimation of the Poverty Incidence and the Magnitude of Poor 

where:
Q = number of families (individuals) with per capita first semester income less than the first semester per capita poverty threshold/line
n  =   total number of families (individuals)

VII. Estimation of Other Foster Greer Thorbecke (FGT) Measures

where:
Z   = first semester per capita poverty threshold/line
Xi  = first semester per capita income of the ith   family
q   = number of families below the first semester poverty threshold

where:
Z   = first semester per capita poverty threshold/line
Xi  = first semester per capita income of the ith   family
q   = number of families below the first semester poverty threshold
n   = total number of families

where:
Z   = first semester per capita poverty threshold/line
Xi  = first semester per capita income of the ith   family
q   = number of families below the first semester poverty threshold
n   = total number of families

¹ Provincial food bundle passed the Test of Revealed Preference