HIGHLIGHTS OF ACCOMPLISHMENTS
I. Improving Productivity and Competitiveness
 

1. Defining Nutritional Status

e. The National Statistics Office (NSO) undertakes household surveys based on a master plan derived from the most recent census of population and housing. The uses of these surveys are for economic and social accounts, program evaluation as well as poverty monitoring and analysis, among others. Two surveys often used for poverty monitoring are the Family Income and Expenditure Survey (FIES) and the Annual Poverty Indicators Survey (APIS). The FIES is on of the basis of the poverty estimates; it includes data on family income and living expenses and related information affecting income and expenditure levels and patterns. In the new millennium, the NSO, FIES included An Analysis of Food Consumption Data in the Philippines for Poverty Estimation.

Study I - An Analysis of Food Poverty Line in the Philippines using the 1993 National Nutrition Survey and the 2000 Family Income and Expenditure Survey, offers an empirical basis in the evaluation of the different methodologies in estimating food poverty line using both the National Nutrition Survey (NNS) and the FIES. The study analyzed the food consumption data in the Philippines using both the 1993 NNS and the 2000 FIES for poverty determination. Inasmuch as the determinations of total poverty line (tpl) and poverty incidence are based on the FIES and the variables needed by the former are the food poverty line (fpl), food expenditure (fe), and total expenditure (te), it is worthwhile to examine the food consumption of the FIES. Furthermore the use of the two data in estimating food poverty line will be studied and compared with the existing method developed by the National Statistical Coordinating Board – Technical Working Group (NSCB-TWG) in order to determine who the poor are.

The FNRI regularly conducts Food Consumption Surveys since the 1960’s, hence, the FNRI’s Food Consumption Survey Results became the “gold standard”. The Institute analyzed the 2000 FIES consumption data by adopting the same system and procedures to  that of the NNS,  and it became the Institute’s significant contribution to the Philippine Statistical System (PSS). The study likewise shed light to the issues regarding the use of either the low-cost menu or the food basket for better estimation of poverty threshold in general.

        Among the four methods studied, the most acceptable and possibly less controversial method is selecting the household belonging to the bottom 30 % income group and then obtaining the intersection of the upper limit of the confidence interval for the mean one-day per capita food consumed among households with less than 100% energy adequacy level, and the lower limit of the confidence interval for the mean one-day per capita food consumed among household with 100% and over energy adequacy. This method is objectively done and is based on the available data to give a more realistic picture of pervasiveness of economic and nutritional deprivation in the country.

        According to the resolution adopted by the National Economic and Development Authority (NEDA), this income group could serve as a basis for monitoring poverty and formulating social welfare assistance programs of the government. The food threshold derived from this group showed how limited income could be utilized to minimize food expenditure and at the same time obtain a nutritious diet. This per capita food threshold could be used as a guide in purchasing a meal that could satisfy energy requirements at very minimal cost. It is emphasized that the food threshold estimate considers only the recommended allowance for energy and did not include those for other nutrients.

        Study II: Estimating Food Threshold and Poverty Incidence Using Food Basket Across Income groups and Bottom 30 % Income Group - The study compared the food basket with food threshold of households across income groups and in the bottom 30 % income group for estimating poverty incidence using per capita consumption of households in the bottom 30 % income class from the 2000 FIES as benchmark indicator.

 

 

Daily per capita food basket derived from “all income group”

 

vs. “bottom 30% income group”: All (urban + rural)

All income group
  
Bottom 30% income group
FOOD ITEMS
Wt. in gms
 
FOOD ITEMS
Wt. in gms
1. Rice
348.0
 1. Rice
407.0*
2. Bread
14.0
 2. Bread
9.7
3. Pork
25.4
 3. Pork
11.1
4. Fish
20.0
 4. Fish
28.5
5. Chicken
10.5
 5. Noodles
3.9
6. Dried Fish
9.5*
 6. Dried Fish
13.7
7. Noodles
6.6
 7. Mungbeans & Products
4.6
8. Mungbeans & Products
5.6
 8. Condiments
13.0**
9. Condiments
16.1
 9. Sugars
15.9
10. Egg
10.9
 10. Fruits
58.0
11. Sugars
20.2
 11. Cooking oil/gata
19.8*
12. Fruits
64.7
 12. Milk
8.8
13. Cooking oil/gata
25.4*
 13. Kangkong/malunggay
24.0**
14. Kangkong/malunggay
24.0**
      kamote & gabi leaves
     kamote & gabi leaves 
 
     
 
Energy (kcal, % RDA)  1994 kcal, 100%
Protein (g, %)50 g, 100%
Vitamin A (mcg RE, %)458 mcg, 100%
Iron (mg %) 12 mg,
80%
                   
*with adjustment in quantity or weight of food item               ** additional

Food poverty line or food threshold is the estimate of minimum income below which a person cannot meet his food needs. The current approach to determine the food poverty line or food threshold is  based on low-cost menu that satisfies 100% of the recommended dietary allowances (RDA) for energy and protein and at least 80% RDA for other nutrients. The rationale of the localized menu-based approach is that it ensures a realistic food threshold from which culturally acceptable dishes or meals may be prepared. However it has the tendency to produce estimates of the food poor, based on different standards of living among regions (or province), whereas the ideal method should have been a standard (food basket) that remains constant among domains. The study addresses  these issues, specifically, the need to identify a food basket from the bottom 30% income group, 10 estimate food threshold and the extent of the population who are likely to be food poor.

 

The policy implications are serious and cannot be ignored. Urban food poverty that was estimated from the food threshold derived form the “all income group” (AIG)  was more than two times (114%) the estimate using the food threshold derived from the “bottom 30% income group.” With regards to food poverty, the implications of the difference in the estimates, the incidence being 9% higher when calculated using “all income groups” vs. “bottom 30% income group,” may not be as disturbing but deserves a closer look as well.

The following observations were made as a result of the study: 1) The food basket of the “all income group” is more diverse and varied, resulting to higher food threshold and food poverty incidence, than that of the “bottom 30% income  group.”   2) Food baskets or lumping of food items into groups vs. actual foods consumed (“unadjusted food basket”) tend to lower the validity (sensitivity and  specificity) of food poverty estimates.  3) The actual food basket consumed by  and which provide the nutritional needs of the households in the bottom 30% income group may be considered by decision-makers in estimating food poverty.  4) The official method was shown to produce inflated estimates of food threshold and food poverty incidence

Steps to Arrive at the Food Basket, Food Threshold and Food Poverty Incidence Estimates

(1)   Aggregating/lumping of similar food items into food groups:

(2)   Shortlist and rank food items based on the distribution of food items by energy and protein contribution in the AIG and bottom 30% income group, urban and rural domains;

(3)   Stepwise non-parametric (kernel) rank discriminant analysis to identify the food items that distinguished households with adequate (100%) intakes for energy and protein from those with inadequate intakes to identify the candidate food basket;

(4)  Stepwise logistic regression, with “having adequate (100%) energy and protein intakes” as dependent variable and the tentative food basket, (in grams), as the independent variables. The food items that came out significant in the model comprised the tentative food basket;

(5)  Adjust the tentative food basket to reach the nutrition standard, which included the addition of foods to meet at least 80% adequacy for vitamin A and iron, and adjust the amount of some foods (e.g. rice, cooking oil) to meet 10% energy, to get the adjusted food basket:

(6)   Compute food cost of the adjusted food basket to estimate food threshold based on the 2000 National Level NSO Price Estimates of various food commodities; and

(7)  Validate the derived food threshold from the 2000 FILES data using food poverty incidence based on total family expenditure of the bottom 30% income group.

 

Copyright © 2008 FNRI-DOST. All Rights Reserved.
  Updated  November 2014
 
DOST Compound, Bicutan, Taguig City
Tel. Nos. (02) 837-2934/837-3164
Best viewed in Mozilla Firefox browser