Philippine Government Statistics on Hunger

EDSA Shangri-La Hotel
October 1-2, 2007
Meeting the Challenge for Official Statistics on Hunger in the Philippines
by
Romulo A. Virola and Lina V. Castro
For additional information, plea se contact:
Author’s name : Romulo A. Virola
Designation : Secretary General
Affiliation : National Statistical Coordination Board
Address : 403 Sen. Gil Puyat Avenue, Makati City
Tel. no. : (0632) 897-2171/ (0632) 895-2395
E-mail : ra.virola@nscb.gov.ph
Co-Author’s name : Lina V. Castro
Designation : Director IV
Affiliation : National Statistical Coordination Board
Address : 403 Sen. Gil Puyat Avenue, Makati City
Tel. no. : (0632) 895-2439
E-mail : lv.castro@nscb.gov.ph
Meeting the Challenge for Official Statistics on Hunger in the Philippines1
by
Romulo A. Virola and Lina V. Castro2
ABSTRACT
The Millennium Development Goal (MDG) under Goal 1, Target 2 declares
to halve between 1990 and 2015, the proportion of people who suffer from hunger. In
line with this, the Philippine Government, through its Accelerated Hunger Mitigation
Program (AHMP), has stepped up efforts to implement focused interventions to
address hunger.
Towards effective monitoring of the MDGs, therefore, there is a need to
assess the impact of these interventions and the current situation of the country in
terms of hunger and food insecurity. While poverty statistics and food poverty
statistics are being generated officially by the National Statistical Coordination Board
(NSCB), statistics on hunger in the Philippines are limited to perception-oriented
household surveys.
Amidst calls for the Philippine Statistical System (PSS) to generate
statistics on hunger, this paper presents a framework for measuring hunger. It
evaluates current approaches and related issues, particularly on the need to
standardize concepts and definitions as well as the need for appropriatelydisaggregated
data.
Keywords: MDG, hunger, Accelerated Hunger Mitigation Program, perceptionoriented
surveys, official statistics
I. Introduction
The Millennium Development Goal (MDG) under Goal 1, Target 2 declares to halve
between 1990 and 2015, the proportion of people who suffer from hunger. In line with this,
the Philippine Government is addressing hunger in the nation’s development agenda
through its Accelerated Hunger Mitigation Program (AHMP), which has stepped up efforts to
implement focused interventions to address hunger.
To be able to monitor the impact of the programs to fight hunger, statistics are
needed. How bad is the current situation? Is hunger rampant throughout the country or is it
clustered in certain areas nationwide? Where do we stand in comparison with other
countries? How do we identify and target the hungry? These are challenges now facing the
Philippine Statistical System (PSS).
The following are some of the existing efforts to generate statistics on hunger,
malnutrition and food insecurity in the Philippines:
1 Paper presented during the 10th National Convention on Statistics, October 1-2, 2007, The authors thank the assistance of
Noel S. Nepomuceno, Jessamyn O. Encarnacion, and Aileen S. Oliveros in the preparation of this paper.
2 Secretary General and Director IV, respectively, of the National Statistical Coordination Board.
National Statistical Coordination Board (NSCB)
Under the System of Designated Statistics3, official poverty statistics including food
poverty statistics are generated by the NSCB. The NSCB releases every three years
provincial estimates of the proportion of those with income insufficient to meet food
requirements, also called subsistence incidence or food poverty incidence or the proportion
of food poor4 individuals/families. Every year, the NSCB also releases food poverty
thresholds.
In addition, the NSCB compiles the Food Balance Sheet of the Philippines which has
hunger-related information, mainly on the food supply and utilization.
National Statistics Office (NSO)
With support from the UNICEF, the NSO has been conducting the Multiple Indicators
Cluster Survey (MICS). In the MICS conducted from February 12 to March 2, 2007 with
2006 as the reference period, one of the modules was on hunger. The survey covered
54,720 individuals in 45,600 households in selected 19 provinces and 5 cities (sampling
error margins of ±3% at the provincial level for selected key indicators), which are pilot
areas for the Sixth Country Programme for Children of the UNICEF. The respondents for
the hunger module were women aged 15-49 years in the sample household. The basic
questions asked were: (1) “Have you experienced not having any meal (breakfast, lunch,
and dinner) in a day in the past week?” (2) If yes, what is your reason? (No money; on
diet; sick; poor appetite; fasting; others (specify). Data to be generated are the incidence of
hunger for the pilot provinces and NCR. The results are expected to be released, by the
NSO this month.
Food and Nutrition Research Institute (FNRI)
The FNRI, thru the National Nutrition Survey (NNS)5 generates percentage of
underweight children 0-5 years old, percentage of underweight adults, prevalence of
thinness among 0-5 year-old children and percentage of stunted children.
3 Per Executive Order No. 352, Series of 1996.
4 Food poor as used here refers to families/individuals with income below the food poverty line, i.e. with income insufficient to
buy the basic food requirements; thus, food poor does not necessarily refer to those whose food consumption is below the
requirement.
5 The NNS is a designated statistical activity programmed to be conducted by FNRI every 5 years. Unfortunately, it has not
always been given top priority by the DOST and the DBM. The NNS was conducted in 1993 and 2003.
In addition, the FNRI administered a module on food security in the 2001 and 2003
NNS to households with 0 to 10-year old children by asking whether the respondent and/or
the respondent’s children have experienced the following situations during the past 6
months, answerable by: (1) never, (2) yes, once during the past 6 months or (3) yes, more
than once during the past 6 months: (a) Knowledge of Self–In the last 6 months, (1) Did you
skip eating or miss meals/food because there was no food or money to buy food? How
frequently did this happen?; (2) Did you ever not eat for a whole day, because there was no
food or money to buy food? How frequently did this happen?; and (3) Were you ever hungry
but did not eat because there was no food or money to buy food? How frequently did this
happen?. (b) Knowledge of Child/Children – In the last 6 months, (1) Did your child/children
skip eating or miss meals/food, because there was no food or money to buy food? How
frequently did this happen? (2) Did your child/children ever not eat for a whole day, because
there was no food or money to buy food? How frequently did this happen? (3) Was/Were
your child/children ever hungry but did not eat because there was no food or money to buy
food? And (c) Knowledge of Situation– In the last 6 months, (1) “I worried that our food
would run out before we got money to buy more”. How frequently was this true?(2) “The food
we bought just did not last and we did not have enough money to get more”. How frequently
was this true? (3) “The children were not eating enough because we did not have enough
food and we could not afford to buy more”. How frequently was this true? (4) “We could not
feed the children nutritionally adequate meals because we do not have enough food and
enough money to buy more”. How frequently was this true? Choices of response for (c) are:
(1) Not true, (2) True, often, and (3) True, sometimes. Although the results were presented
during the seminar series/dissemination forum of the NNS in July and December 2004, the
figures were not published.
Department of Agriculture/Bureau of Agriculture Statistics (DA/BAS)
The Bureau of Agricultural Statistics conducted the Survey of Hunger Incidence in
the Philippines in August 2006, with April to June 2006 as the reference period, covering a
total of 13,400 sample households in 1,340 sample barangays in 78 provinces, two
chartered cities and NCR. The respondent was any responsible adult member of the sample
household. The questions asked were: (1) “During April to June 2006, did it happen even
once that your household experienced hunger and have nothing to eat?” (2) “If yes, how
often? (Once; a few times; often; and always). (3) What was the main reason why your
household experienced hunger? (food available but cannot afford to buy; food not available
although can afford to buy; food not available and cannot afford to buy; residence far from
the source of food; and others (specify). Data generated from the survey include: proportions
of households that experienced moderate and severe hunger by region and province; and
percentage distribution of households that experienced hunger by reason by region and
province. The questions are obviously patterned after and duplicative of the SWS survey to
be described later, but unlike the SWS or the FNRI surveys, the BAS survey produces
provincial level estimates.
National Nutrition Council
The NNC has the following frameworks/indicator systems on malnutrition, food
insecurity and hunger:
¨ Operation Timbang (OPT) – The OPT is an annual activity of the NNC in order to
assess undernutrition at the local level. It involves weighing of all preschoolers 0 –71
months old or below six years old in a community to identify and locate the
malnourished children. It uses weight-for-age as indicator in assessing protein-energy
malnutrition among preschool children. From the processed OPT results, a list of
nutritionally depressed cities/municipalities is generated for dissemination to
government and non-government organizations for prioritization in nutrition
programming and interventions.
¨ Local Nutrition Early Warning System(LNEWS) – The LNEWS provides information on
sudden or impending deterioration of the nutrition situation in the community. It was
pilot tested by the NNC in selected municipalities in Iloilo and Ifugao and included
seven indicators as follows: price of staple food and fish; frequency of consumption of
staple food; frequency of consumption of vitamin A-rich foods by children < 7 years
old; prevalence of underweight preschool children; incidence of low birth weight;
morbidity incidence; and occurrence of weather and environmental disturbance.
Social Weather Station (SWS)
The private sector in the Philippines has been very active in generating statistics in
addition to those coming from government agencies. The SWS generates statistics on
hunger based on perception -oriented household surveys. The SWS survey on hunger is
conducted every quarter with a sample size of 1,200 households divided into samples of 300
each in the National Capital Region (NCR); balance of Luzon; Visayas; and Mindanao
(sampling error margins of ±3% for national percentages and ±6% for area percentages).
The respondents are the household heads and different sets of respondents are selected
per quarter. The questions asked are: “In the last three (3) months, did it happen even once
that your family experienced hunger and did not have anything to eat?”. “If yes, did it happen
only once, a few times, often or always?” Data generated every quarter are the incidence of
hunger by location and degree of hunger.
Food and Agriculture Organization (FAO)
· At the international level, the FAO has developed the Food Insecurity and
Vulnerability Information and Mapping System FIVIMS) which aims to identify
and characterize people who are at risk to food insecurity and vulnerable to
malnutrition. The NNC serves as the Philippine focal agency for FIVIMS. As
implemented by the NNC, FIVIMS comprises 12 core indicators, which are
measures of either the cause or effect of food insecurity and vulnerability as
follows: percent of food expenditure/total expenditure; percent of cereal
expenditure/total food expenditure; poverty incidence; median family income;
ratio of per capita expenditure/per capita income; percent of households with
access to safe water; percent of agricultural lands under tenancy; percent of
underweight children 0-5 years old; percent of underweight adults with BMI <
18.5 kg/m2; percent of families with working children 5 –17 years old; cohort
survival rate for elementary level; and unemployment rate. These core
indicators, which are sourced from various government surveys and reports,
are used to compute a composite index of food insecurity and vulnerability.
The country’s provinces were then grouped into 5 clusters based on the
composite index as follows: (1) Not Vulnerable; (2) Less Vulnerable; (3)
Vulnerable; (4) Very Vulnerable; and (5) Very Very Vulnerable.
· A number of food security statistics are also available on the FAO website at
http://www.fao.org such as (1) food supply measured in
kilocalories/pe rson/day, (2) number of undernourished and (3) proportion of
undernourishment. The FAO measure of food deprivation, referred as the
prevalence of undernourishment, is based on a comparison of usual food
consumption expressed in terms of dietary energy (Kcal) with minimum
energy requirement norms. The part of the population with food consumption
below the minimum energy requirement is considered underfed (FAO 2003).
International Food Policy Research Institute (IFPRI)
In addition to the FIVIMS index of the FAO, the IFPRI compiles the Global Hunger
Index (GHI) based on three equally weighted indicators: the proportion of undernourished as
a percentage of the population (reflecting the share of the population with insufficient dietary
energy intake); the prevalence of underweight in children under the age of five (indicating the
proportion of children suffering from weight loss and/or reduced growth); and the under-five
mortality rate (partially reflecting the fatal synergy between inadequate dietary intake and
unhealthy environments). By combining the proportion of undernourished in the population
with the two indicators relating to children under five, the IFPRI hopes to ensure that both the
food supply situation of the population and the effects of inadequate nutrition on a
physiologically very vulnerable group are captured.
Now, what do the statistics from these various sources tell us? Do they validate or
contradict each other?
· Latest NSCB (2006a) official estimates show that in 2003, about 10 out of 100
families and 14 out of 100 individuals were food poor.
· On the other hand, official data from the FNRI (2006) estimate that out of 100
children 0-5 years old, about 25 are underweight (suffering from current
malnutrition), 26 are stunted (chronic malnutrition) and 5 are wasted (acute
malnutrition); moreover, 56.9% of households did not meet the recommended
nutrient energy intake/ requirement in 2003.
· Findings from the same nutrition surveys of the FNRI cite that 84.4% of the
households with 0-10 year old children experienced food insecurity in 2001
and 77% in 2003.
· The latest perception-oriented survey of the SWS (2007) with the second
quarter of 2006 as reference period estimates the proportion6 of overall
hunger at 19.2%, moderate hunger at 15.2% and severe hunger at 4.0%.
· The latest list of 100 nutritionally depressed municipalities based on the OPT
for 2004, as released by the NNC, included 34 municipalities in the Bicol
Region, 19 municipalities in Eastern Visayas, 17 municipalities in Western
Visayas, 8 municipalities in Central Visayas, 5 municipalities both in Caraga
and CALABARZON, 4 municipalities in SOCCSKSARGEN, 3 municipalities in
Davao region, 2 municipalities both in Cagayan Valley and MIMAROPA, and
1 municipality in Northern Mindanao. Note that municipalities in ARMM were
excluded because their OPT coverage were less than 80% or results were
not submitted at all.
· Results of the BAS one time survey conducted in August 2006 estimate that
from April to June 20067, about 18.6 percent of the households in the country,
6 Having experienced hunger during the past three months once is overall hunger, a few times is moderate hunger
and many times/frequently is severe hunger.
experienced hunger. Of this, 15.0 percent experienced moderate hunger and
3.6 percent, severe hunger.
· Latest results of the Philippine FIVIMS as implemented by the NNC, estimate
that food insecurity in the Philippines is prevalent in 49 provinces in varying
degrees as follows: (1) 38 provinces as Vulnerable; (2) 8 provinces as Very
Vulnerable; and (5) 3 provinces as Very- Very Vulnerable.
· The resulting Global Hunger Index (GHI) as computed by the IFRI was
expressed as a percentage, varying between 0 and 100, where low scores
indicate less hunger and high scores indicate severe hunger with 4 categories
as follows: (1) Low to Moderate, (2) Serious, (3) Alarming, and (4) Extremely
Alarming. The GHI for the Philippines at 17.6 for 2003, p laced the country in a
serious hunger situation together with Indonesia, Thailand and Vietnam at
12.5, 12.4 and 18.4 respectively. Malaysia at 7.2 is considered in a low to
moderate hunger situation, while Bangladesh, India and Pakistan at 28.3,
25.8 and 21.8 respectively, are placed at alarming situations.
· From a table culled from the FAO website on food security (Virola, 2006) the
following findings are shown: (1) In terms of food availability, in 1990-92, the
Philippines 2260 kcal/person/day, was ahead of Thailand, Vietnam, Lao PDR
and Cambodia at 2250, 2180,2110 and 1870 kcal/person/day, respectively.
By 2000-02, Vietnam and Thailand at 2530 and 2450 kcal/person/day
respectively, had caught up with the Philippines at 2380 kcal/person/day and
Lao PDR at 2290 kcal/person/day is getting close. (2) In 1990-92, there were
more undernourished Pinoys (16.2 million), than other Southeast Asians
except the Vietnamese (20.6 M) and the Indonesians (16.4M). By 2000-02,
undernourished Pinoys (17.2M) were the biggest group of undernourished
Southeast Asians. (3) In all three periods, the incidence of undernourishment
is higher in the Philippines than for Southeast Asia as a whole and for all Asia
and the Pacific. In 1990 -92, the proportion of undernourishment in the
Philippines at 10% was lower than in Cambodia, Vietnam, Lao PDR and
Thailand at 43%, 31%, 29% and 28%, respectively. By 2000 -02, only
Cambodia at 33% had a higher proportion of the undernourished than the
Philippines at 22%.
A summary of these statistics as presented in Table 1 point to the fact that there are
a number of measurement activities or data systems that exist at the local, regional, national
7 The BAS and SWS surveys cited have the same reference period: 2nd quarter of 2006.
and even at the global levels, with each one providing different estimates based on various
concepts, varying target population and various methods at different levels of
disaggregation. The figures do not seem to validate each other. For example, the following
are estimates from three different sources (survey-based): (1) in 2003, 77.0 % of households
(with 0-10 children) experienced food insecurity (FNRI, 2003 NNS); (2) BAS reports that
13.9% of households experienced hunger in the second quarter of 2006 (BAS, 2006 SHIP);
and (3) SWS reports an overall hunger of 18.6% for the second quarter 2006. We should
examine and assess whether the fluctuations between the 2003 and 2006 estimates are
“acceptable” results of valid measures of hunger. Is the difference of 5 percentage points in
hunger at the national level (BAS vs. SWS) “acceptable” as results of valid, reliable,
measures of hunger? One thing is certain, there remains no coordinated plan for an overall
monitoring of the hunger situation at the national, regional and provincial levels, and even
down to the municipal level. At this point, it is important to consider the broader goals of the
Accelerated Hunger Mitigation Program of the government that should address hunger in a
holistic manner that should provide what dimensions and what degrees of severity of “food
insecurity” or hunger are most relevant to measure and monitor not only at the provincial
level, but at the municipal level as well. On the supply side, measures could be along
producing more food and efficient delivery of foods to whom and where it is needed. On the
demand side, measures could be along putting more money in people’s pockets and in
promoting good nutrition among others. Within this perspective, should the measures then
be limited to the unique individual experience in the direct sense of the uneasy or painful
sensation of not having enough to eat? Or to a notion of hunger that arises in the context of
inadequate access of individuals to sufficient, and nutritious food both in quantity and quality
to meet their dietary requirements for a healthy and productive life? In other words,
measuring the “causes” and “effects” of hunger as well?
Obviously, it is desirable to come up with a statistical development program for an
improved and rational measurement of hunger. Given the limited resources of government
and of the PSS, given the hunger-related statistics that already exist in the country, is it
necessary to propose/conduct new statistical surveys/censuses/administrative -based record
systems? Is it wise for government agencies like the BAS and the FNRI to be spending
resources on ba sically the same survey on hunger? Is it prudent for government to duplicate
what the private sector like the SWS is doing? True, these different sets of statistics may be
serving different and valid purposes but is the value added commensurate with the co st? No
doubt, we need to prioritize and rationalize statistical activities on hunger in order to
maximize the use of our very limited resources. However, there are some considerations in
generating official statistics that we should address as well. In terms of relevance, we should
fully understand what our users really need. Are we responding to their needs? Do users
translate into policies the statistics they say they need? Can the statistics help define policies
and programs? Do the statistical offices have the resources needed, both manpower and
financial? What are our obligations to our citizens – Is it something we should do? We
should also consider the response burden in view of the similar surveys conducted on the
same subject matter. What is the value added of new statistics generation initiatives? Will it
not create an information overload?
Toward this end, the paper presents a proposed conceptual and operational
framework in the following section. The third section discusses the preliminary results and
feedback from stakeholders on the framework. In the last section, some concluding remarks
and recommendations are presented.
II. Meeting New Challenges for Official Hunger Statistics: Proposed Hunger Index
Until 2001 with the FNRI module on food security in the NNS, the PSS was not
addressing directly the measurement of hunger. Hunger data from the perception-oriented
surveys of SWS, which report percentage of households having “experienced hunger, with
nothing to eat, at least once in the past three months” are definitely much easier to collect
than the usual economic and social data collected by the PSS. However, they are subject to
issues of validity, objectivity and transparency. Do they capture sentiments and perceptions
beyond hunger, such as for instance, dissatisfaction with the government and its leaders?
In early 2006, the President issued a series of Cabinet directives to hasten the
implementation of government measures to alleviate poverty, specifically on the need to
rationalize government programs and implement more focused interventions to address
hunger. A Cabinet memorandum8 directed the “National Nutrition Council (NNC), National
Economic and Development Authority (NEDA), NSCB, Bureau of Agricultural Statistics
(BAS), and the FNRI” to establish benchmarks/index on hunger. In response, the NSCB
(2006b) created an inter-agency Task Force on the Development of Hunger Index (TF) to
generate quick proxy indicators on hunger; develop a methodology to generate hunger index
and establish benchmark information on hunger. The NSCB (2006c) TF reviewed various
conceptual frameworks and possible data sources on hunger, malnutrition and food
insecurity and looked for alternative approaches. Subsequent activities however, required
capacity building to include resources such as manpower and logistics for framework
8 Cabinet Memorandum dated 17 January 2006, “Implementation of Focused Interventions against Hunger”
development, conduct of validation and user-consultation workshops as well as technical
expertise on the construction of the hunger index. Thru multi-agency initiatives, the FAO
provided funds for the Project “Establishment of Benchmark Data and Index on Hunger”
aimed: (1) to provide data support for more focused interventions to address hunger by (a)
formulating a methodology for the construction of a hunger index based on existing data;
and (b) establishing benchmark hunger index; and (2) identify and recommend statistical
policies for the institutionalization of data generation for the hunger index.
Concepts and Definitions
The FAO (2000) defines hunger as a condition in which people do not get enough
food to provide the nutrients for fully productive, active, and healthy lives. Food insecurity, on
the other hand, is a condition characterized by the lack or absence of available, adequate,
accessible, affordable, safe, and nutritious foods that satisfy the dietary needs and food
preferences of all people at all times for an active and healthy life. Related terms are
malnutrition, undernutrition and undernourishment. How do we measure these observable
facts? Statistically speaking, a social phenomenon like hunger is not as easy to measure as
some economic variables like employment and prices. Hunger, like poverty, is multidimensional
in nature; thus, capturing its many facets would require a sound framework
scientifically grounded on quality statistics.
The Proposed Conceptual Framework of the NSCB FAO-funded Project
Inadequate food consumption and poor health/nutritional status are two obvious
correlates of hunger that may lead to the most tragic consequence of hunger, which is early
death. These are caused by intermediate factors, namely, unavailability and inaccessibility of
food, ignorance and other socio-economic factors, inadequate health care and health
services, and demographic factors such as overpopulation, all operating independently and
interactively. According to the United Nations (UN) MDG Report (2005), “Malnutrition in
children contributes to over half of child deaths. It is caused not only by food deprivation, but
also by the debilitating effects of infectious diseases and lack of care. Strategies to combat
child malnutrition include exclusive breastfeeding for the first six months, increasing the use
of micronutrient supplements, reducing infectious diseases, and improving access to clean
water and sanitation.” These may be caused by improper economic policies and political
infrastructure, low level of technology, negative cultural and demographic forces, and
underdeveloped natural resources.
The NSCB TF decided to adopt the FAO definition of hunger. Using this definition,
Florentino (2007), proposed a framework (Figure 1) encompassing the essential dimensions
of hunger – its immediate, intermediate and basic causes/correlates and its ultimate result.
Thus, the proposed framework recognizes the multi-dimensional nature of hunger that goes
beyond the physical sensation associated with lack of food intake. It also takes into
pragmatic consideration the limited manpower and financial resources of government and of
the PSS to undertake new primary data collection activities to measure hunger.
Components of the Proposed Hunger Index
Based on the proposed framework on hunger index, inadequate food consumption
and poor health/nutritional status are identified as the immediate causes of hunger.
Accordingly, the following are the proxy indicators identified to come up with the hunger
index: (a) proportion of households with per capita energy consumption less than per capita
energy requirement for inadequate food consumption, and (b) proportion of underweight
children under 5 years old, and (c) mortality rate of children under 5 years old for poor
health/nutritional status. Data were assessed in terms of the possible sources, availability
starting 1990 and onwards, and level of disaggregation:
1.Proportion of households with per capita energy consumption less than the requirement –
mainly sourced from the Food Consumption Survey (FCS) module of the NNS conducted by
the FNRI every five years. The FCS obtains actual amounts of food consumed in the
household and outside the home for one day using the food weighing technique. The
adequacy of the energy and nutrient intakes is assessed against the Recommended Energy
Figure 1. Conceptual Framework of Hunger
HUNGER
Early Death
Inadequate Food Consumption Poor Health/Nutritional Status
Unavailabilit
y of Food
Inaccessibilit
y of Food
Ignorance
& other
Sociocultural
Inadequate Health
Care & Health
Services &
Unhealthy
Overpopulation
&
other
demographic
Factors
Economy &
Technology
Policy & Culture
Ecology and
Natural Resources
Source: Proposed Hunger Index for the Philippines, Conceptual Framework and Indicators, Report of Consultancy.RF Florentino, 2007
and Nutrient Intakes (RENI). The last NNS conducted was in 2003. Although the NNS is
undertaken every five years, the FCS was included only in the years 1993 and 2003.
Unfortunately, the 2003 NNS was designed to generate national and regional estimates only.
In its efforts to enhance the relevance of the NNS, the FNRI provided provincial level
estimates with corresponding coefficients of variation (CVs) as well as margin of errors
(MEs) to guide the users on the reliability of the data. Data from the 1993 NNS are thus
available at the national, regional and provincial levels.
2.Proportion of underweight children under 5 years old – mainly sourced from the module on
Anthropometric Nutrition Survey of the NNS. This survey involves the measurement of
height, weight, skinfold thickness, waist/hip and mid -upper arm circumferences of children
aged 0 to 19. In between NNS years, the FNRI conducts the Updating of the Nutritional
Status of Filipino Children, specifically for anthropometric data. Thus, the proportion of
underweight children under 5 years old is available every two or three years. However, data
for 2003 are available only at the national and regional levels As in the data on energy
consumption, the FNRI performed special runs to generate, for purposes of the FAO project,
provincial data on underweight children with the corresponding coefficients of variation (CVs)
as well as margin of errors (MEs).
Note that one possible source of data on underweight children down to the municipal
level is the OPT of the NNC which is conducted in practically all municipalities in the country.
However, the quality of data generated needs further study considering the various issues
raised such as the use of a nonstandardized/noncalibrated weighing scales and the low
coverage in some areas.
3.Mortality rate of children under 5 years old – The National Demographic and Health Survey
(NDHS) of the NSO is the main source of data on mortality rate. The NDHS is conducted
every five years with the latest in 2003. The NDHS contains the estimates on early childhood
mortality rates by neonatal, postnatal, infant, child and under-five mortality rates, early
childhood mortality rates by socio-economic characteristics and region, by demographic
characteristics, and by women’s status indicators. The data on the mortality rates are direct
estimates from the questions in the reproductive history section of the Women’s
Questionnaire.
The data available for Under-5 Mortality Rate (U5MR) from the 2003 NDHS are only
at the regional level. Although mortality data can also be sourced from the civil registration
records of the NSO, there could be underestimation of deaths considering the under
registration of deaths in civil registration systems. Nevertheless, the data from the civil
registration records were used to come up with provincial data for 2003. The data are
adjusted for under registration and pro -rated using the structure of the Regional U5MR data
from NDHS9. On the other hand, indirect estimates of U5MR from 1990 – 1995 are available
up to the provincial level, while the1998 and 2003 data points are only available at the
regional level owing to sampling variability or the smallness of the sample due to rarity of the
event.
Chart 1summarizes the availability of data for the proposed hunger index.
Chart 1. Summary of Hunger Components and Indicators
Component Indicator/ Data Items Main
Data Source
Years
Available
Level of
Disaggregation
Inadequate food
consumption
Proportion of households
with per capita energy intake
< RENI
National Nutrition Survey,
FNRI 1993, 2003
Regional/
Provincial
Poor health/
nutritional status
Proportion of Underweight
Children under
5 yrs
National Nutrition Survey,
FNRI
1990,1992,
1993,1996,1998,
2001, 2003, 2005
Regional/ Provincial
Under 5 Mortality Rate
National Demographic and
Health Survey
NSO
1990-1995,1998,
2003
1990-95: Provincial
1998 & 2003:
Regional
Source: Data Assessment Report, NSCB, 2007
Proposed Estimation Methodology
From the framework, an index meant to capture the depth and extent of hunger is
defined with the operational framework shown in Figure 2. The Hunger index (HI) is defined
as a measure of severity of hunger in the population. It is computed using proxy indicators,
namely, proportion of households with per capita energy consumption less than the
requirement (E), proportion of underweight children under 5 years (U) and mortality rate of
children under 5 years (M). The indicators are equally weighted to come up with the index.
The Index is from 0 to 1, with higher scores depicting greater severity of hunger. Thus,
HI = (E + U + M)/3
The proposed hunger index gives the same weight to information given by its three
component indicators. An alternative index developed by the Statistics Consultant10 is one
9 These figures were pro rated to be consistent with the Regional Under 5 Mortality Rate from NDHS using the following
formula: PA = (PNSO) x (RNDHS)
RNSO
Where: PNSO = Provincial U5MR from NSO, RNDHS = Regional U5MR from NDHS
RNSO = Regional U5MR from NSO, PA = Provincial Adjusted U5MR
10 Done by Dr. Lisa Grace S. Bersales using the conceptual framework developed by Dr. Rodolfo Florentino.
with unequal weights developed using factor analysis. Adjectival hunger ratings from the
values of the hunger index; i.e., alarming, serious, moderate or low were also developed
using the medium-term Philippine Development Plan targets for the three indicators as
shown below. The correlation of the proposed HI with each of the following indicators was
likewise computed and tested for significance: FIVIMS; subsistence incidence; and
magnitude of the poor population.
Source: Proposed Hunger Index for the Philippines, Conceptual Framework and Indicators, Report of Consultancy, RF
Florentino, 2007
Chart 2. Proposed Hunger Index Categories
CATEGORY LOWER VALUE UPPER VALUE
Low 0.000 0.256
Moderate 0.257 0.291
Serious 0.292 0.326
Alarming 0.327 1.000
Source: The Development of a Hunger Index for the Philippines: Methodology and Results . LGBersales, (2007)
Figure 2. Operational Framework of Hunger
HUNGER
Inadequate Food Consumption Poor Health/Nutritional Status
Proportion of underweight children
under 5 yrs of age
Under 5 Mortality Rate
Key
Indicators
Proportion of HH with per
capita energy consumption <
requirement
III. Preliminary Results: What Do the Stakeholders Say?
The NSCB (2007) in coordination with the Statistics Consultant, came up with
preliminary estimates of the hunger index at the national level, by region and province for
2003, using as data sources the following: FNRI (2006) for E and U and NSO (2004) for M.
The proposed hunger index was estimated at 0.272 which indicates moderate hunger
for the Philippines. Among regions, hunger is categorized serious in ARMM, Zamboanga
Peninsula, Central Visayas and Eastern Visayas, while regions classified with low severity of
hunger include CAR, NCR, Davao Region and Central Luzon. Note that no region is
classified with an alarming hunger situation. Comparing the ranks of hunger index with
poverty incidence, the most hungry regions, ARMM and Zamboanga Peninsula are ranked
the 2nd and 3rd poorest, respectively.
Knowing that there is some degree of concordance 11 with the regional ranks in terms
of poverty incidence, it is worthwhile to go down to the provincial level estimates.
Table 3 presents hunger index by region and province. Eighteen (18) provinces are
tagged with alarming hunger conditions, 15 provinces – serious, 20 provinces – moderate,
and 30 provinces – low hunger conditions. The 6 most alarming provinces include 2
provinces in ARMM and one each in Regions 8,10, 12 and MIMAROPA as follows: Cotabato
City; Northern Samar; Sulu; Romblon; Misamis Occidental; and Tawi-tawi. The 6 provinces
with the least hunger severity are found in Region 6, CAR, Region 3, MIMAROPA and
Region 10: Guimaras; Ifugao; Bataan; Occidental Mindoro; Apayao; and Camiguin.
Table 4 shows the estimates of the hunger index by region and province with the
corresponding component indicators. Additional correlation tests performed on the hunger
index showed significant positive correlation between the proposed provincial HI and
subsistence incidence (r = 0.29, a = 0.01) as well as regional HI with the magnitude of the
poor population (r = 0.57, a =0.02).
What do the stakeholders say?
A Users’ Forum was conducted by the NSCB last July 31, 2007 to solicit comments
and recommendations on the proposed conceptual framework and the preliminary
11 Kendall’s tau-b rank correlation coefficient of 0.426 is statistically significant at 5 % level.
estimates. The users welcomed the concerted effort in the establishment of a hunger index,
for monitoring the country’s overall goal of food and nutrition security for all Filipinos, but
reservations and even opposition to the proposed hunger index were expressed. More
specifically, the comments included the following:
(1) From the local government units (LGUs) perspective – The resulting
statistics are critical as basis for the LGUs in undertaking various
development activities and ultimately enable them to update/revise
existing policies and to formulate more strategic policies aimed at better
utilization of resources and targeting of programs and project
beneficiaries.
(2) From the academe and other users – The need to perform specificity and
sensitivity tests. The hunger measure should provide direct answers or
information on: Who are the hungry? How can they best be identified and
characterized? Where are the hungry? The requirement on the use of
standardized method for identifying the hungry across population groups
may not be appropriate as determinants of hunger may vary from one
place to another and across socio-economic groups. The “proxy indicator”,
proportion of households with per capita energy intake below the
requirement, can be actually treated as a direct indicator, but could be
subject to measurement errors. There is a need for more frequent
monitoring of hunger because the country experiences disasters of
different types and magnitude. The significant correlation of the hunger
index with the existing official statistics like the subsistence index may
mean that there is no need for another measure. Consider other
approaches in formulating the hunger index.
These comments were surely made for the PSS to have a good and useful measure
of hunger. In assessing the comments it would help if alternative approaches are offered or
identified. The competing approaches could then be examined and compared in terms of
cost and benefits.
Without doubt, the PSS, and the NSCB in particular, is interested in providing good
measures of hunger in the Philippines on a regular basis. The wish list could include not only
the prevalence of hunger among individuals/households/families living in different areas or
even belonging to different socio-economic groups, but also the identification of the hungry.
However, as envisioned by the NSCB TF, the hunger index is meant to be a
conceptually-valid but practical tool to assess the level or severity of the country’s hunger
problem and to measure progress over time in overcoming it. It was conceptualized to
incorporate aspects that reflect the multidimensional nature of hunger, be they “causes” or
“effects”. It is based on a framework that examines the “availability and utilization of food”
and the way food is “converted” into healthy bodies and healthy lives. Nobel-winning
economist Amartya Sen (1983) argues for a “capability approach” to hunger that treats
calories as a means to the more important ends of “being well-fed” and “being in good
health”. On the other hand, the proposed hunger index goes beyond the measurement of
calories. In other words, in addressing the hunger problem, we should produce hunger
statistics that reflect not only on how food is converted into human functioning and capability
(good health) but also indicate measures of hunger beyond availability and access to
utilization. Thus, a broader understanding of hunger entails more than simply looking at the
calorie or energy supply. The broader perspective of hunger is trying to account how food is
used and balanced to avoid malnutrition and promote well- being.
Moreover, the proposed HI was formulated with due consideration to resource
constraints that may not allow new primary data collection activities; thus, the proposed HI
uses only existing or available data.
The proposed HI can identify which areas (regions/provinces) are in alarming
conditions and therefore need priority attention. Unlike other measures, HI has easy
interpretation in the sense that HI scores moving towards 1 are indications of setbacks in the
fight against hunger. By examining the components of the HI, policy-makers can identify
aspects of hunger that need greater attention. Insights into the more specific causes of
inadequate food consumption and poor health status in the alarming areas can be gained
through a closer analysis of other socio-economic data in these communities. Finally, the HI
can be used by the government for targeting purposes under its Accelerated Hunger
Mitigation Program.
IV. Concluding Remarks and Recommendations
While the initial efforts of the PSS have made some progress towards a quantitative
measurement of hunger, much remains to be done.
In addition to the comments raised by the stakeholders during the consultative forum,
the validity of the index as well as the quality of the data sources, including timeliness and
use of standard concepts and definitions should be examined more thoroughly. Some
administrative-based records like those on morbidity and mortality are unreliable because
events are recorded where health facilities are located, generally in economic centers. This
results in the overestimation of the severity of a problem in the economic centers. Also,
survey-based indicators like E, U and M are not reliably generated below the regional level.
It is unfortunate that the NSO surveys, which used to generate provincial level estimates no
longer do so, because of concerns about unacceptable coefficients of variation12. In
addition, the importance of the timeliness13 of the release of the hunger index should be
recognized and addressed.
A follow-up effort being proposed as Phase 2 of the FAO-assisted project is the
development of a methodology for the construction of a hunger index based on existing
and/or new household data. This would entail the design of a module on hunger in an
existing household survey, e.g., through the Labor Force Survey of the NSO, which can
provide direct estimates of hunger prevalence and count of the hungry.
Validation workshops and more consultation fora with stakeholders should be
conducted further. Advocacy activities on the policy uses of the hunger index should likewise
be undertaken. The proposed methodology and estimates are also scheduled for discussion
at the NNC Technical Committee and NNC Governing Board which are the policy bodies
that oversee the implementation of measures against hunger and which are expected to be
the major users of hunger statistics. Subsequently, statistical policies will have to be
formulated for approval by the NSCB Executive Board for the institutionalization of data
support for the generation of the hunger index.
Towards enhanced policy relevance of the hunger index, it should be generated with
the lowest possible geographical dissagregation, provincial at the minimum, but taking into
consideration resource constraints. In this regard, the existing efforts of the PSS in the use
of small area estimation techniques will be useful.
With the limited resources available, it is imperative that duplicative and overlapping
statistics generation efforts be avoided. Finally, the institutionalization of the hunger index in
the PSS may take some time, but present efforts should certainly pave the way for the
generation of official statistics on hunger in the Philippines. For this to happen, government
as well as the private sector must recognize and appreciate the need to invest in statistics.
12 The Australian Bureau of Statistics releases information with CVs higher than 10%, with the necessary caveats.
13 For the computations done in the project, the most timely estimate possible using available data is for 2003.
References
BAS (2007). Survey of Hunger Incidence in the Philippines (SHIP).
Bersales, L.G.B. (2007). The Development of a Hunger Index for the Philippines: Methodology and
Results
FAO, 2000. The State of Food Insecurity in the World http://www.fivims.net
FAO, 2003. FAO Methodology for the Measurement of Food Deprivation, FAO Rome, October 2003.
Florentino, R.F. (2007). Proposed Hunger Index for the Philippines, Report of Consultancy.
FNRI, 2006. Philippine Nutrition, Facts and Figures 2003. 6 th National Nutrition Survey 2003.
FNRI, 2006. 2005 Updating of Nutritional Status of Children. Terminal Report.
IFRI, 2006 Global hunger Index: A Basis for Cross Country Comparison, October 2006.
NNC, “Identifying Food Insecure and Vulnerable Areas in the Philippines through FIVIMS”, NNC
Report to the Cabinet, 17 January 2006
NSCB, 2006a, http://www.nscb.gov.ph. June 6, 2006
NSCB, 2006b. NSCB Memorandum Order No. 001 Series of 2006.
NSCB, 2006c. Preliminary Report on Proxy Indicators on Hunger. Task Force on the Development of
Hunger Index.
NSCB, 2007. Steering Committee Meeting, Establishment of Benchmark Data and Hunger Index
Project
NSO, 2004. 2003 National Demographic and Health Survey.
SWS, 2007. http://www.sws.
Sen, A., “Family and Food: Sex Bias in Poverty,” in “Resources, Values and Development”,
(Cambridge, MA: Harvard University Press, 1983).
United Nations Millennium Development Goals Report 2005, UN, New York, 2005
Virola, R.A., Statistically Speaking, “Gutom Ka Ba?”, http://www.nscb.gov.ph, March 2006.
Various Sources: NNC, SWS, FNRI, NSO, FAO, IFPRI, NSCB and BAS
Table 1. Current Approaches in Measuring Hunger in the Philippines
Agency Methodology Indicator Latest Estimate/s
Reference
Period Frequency
Coverage/
Disaggregation
FIVIMS Classification of provinces by
vulnerability to food insecurity
38 provinces – V; 8
provinces – VV and 3
provinces – VVV
2003 Not specified Provincial
percentage of preschoolers below
six years old who are
malnourished
no aggregation at national
level
2004 Annual Municipal
List of Nutritionally Depressed
Cities/ Municipalities(NDM)
First 100 NDMs 2004 Annual Municipal
SWS Survey on
Poverty and
Hunger
percentage of reporting
experience of hunger in the past 3
months
19.0% – overall hunger
15% – moderate hunger
4% – severe hunger
First Quarter
2007
Quarterly National and by island
groups
Percentage of underweight,
stunted, wasted – 0-5 children
underweight – 25.0%;
underheight –
26.0%;wasted – 5.0
2006 every 5 years, with
updates every 2 – 3
years
National, Regional
Percent of HHs w/ less 100%
energy adequacy
Phil. – 56.9% 2003 every 5 years, with
updates every 2 – 3
years
National, Regional
Percent of HHs w/ 0-10 children
that experienced food insecurity
77.00% 2003 done only in 2001
1nd 2003
National, regional
NSO MICS percentage of HHs reporting not
having any meal in a day
no figures yet 2006 one time survey
module
NCR and 19 selected
provinces and 5
municipalities
FAO FAO
Methodology for
the
Measurement
of food
deprivation
proportion of undernourished Phil. – 22% 2002 -2003 covers 126 countries
IFPRI Global Hunger
Index
Hunger index Phil.- 17.55 2003 covers 116 countries
NSCB Poverty and
related
indicators
Subsistence incidence 10% of families – food
poor;14% of total
population – food poor
2003 every 3 years national, regional,
provincial
BAS SHIP percentage of HHs reporting
experience of hunger in
Phil.: experienced hunger –
18.6%; moderate hunger –
15%; severe hunger – 3.6%
2nd Quarter
2006
one time survey national, regional,
provincial
NNC
OPT
FNRI NNS
Table 2. Hunger index estimates, degree of severity and ranks by region, 2003, with
Poverty Incidence (PI), families
Region HI Severity Rank PI Rank
Philippines 0.272 Moderate 24.40%
National Capital Region
(NCR) 0.233
Low
16
4.8 17
CAR 0.222 Low 17 25.8 11
Ilocos 0.273 Moderate 10 24.4 12
Cagayan Valley 0.264 Moderate 13 19.3 14
Central Luzon 0.247 Low 14 13.4 16
CALABARZON 0.275 Moderate 9 14.5 15
MIMAROPA 0.27 Moderate 12 39.9 5
Bicol 0.293 Moderate 6 40.6 4
Western Visayas 0.279 Moderate 8 31.4 9
Central Visayas 0.305 Serious 4 23.6 13
Eastern Visayas 0.305 Serious 3 35.3 7
Zamboanga Peninsula 0.318 Serious 2 44 3
Northern Mindanao 0.281 Moderate 7 37.7 6
Davao Region 0.236 Low 15 28.5 10
SOCCSKSARGEN 0.299 Moderate 5 32.1 8
Caraga 0.271 Moderate 11 47.1 1
ARMM 0.323 Serious 1 45.4 2
Source: The development of a Hunger Index for the Philippines: Methodology and Results
Table 3. Hunger Index by Region and Province and by category, 2003
Region/ Province
Hunger
Index
HI Rank Severity HI Rank
PHILIPPINES 0.272 Moderate
Region I 0.273 10 Moderate 10
Ilocos Norte 0.269 45 Moderate 42
Ilocos Sur 0.283 36 Moderate 38
La Union 0.295 31 Serious 31
Pangasinan 0.265 49 Moderate 47
Region II 0.264 13 Moderate 13
Batanes* – – – –
Cagayan 0.281 38 Moderate 33
Isabela 0.251 58 Low 57
N. Vizcaya 0.282 37 Moderate 37
Quirino 0.231 67 Low 70
Region III 0.247 14 Low 14
Bataan 0.178 81 Low 81
Bulacan 0.236 65 Low 66
N. Ecija 0.251 57 Low 60
Pampanga 0.267 47 Moderate 49
Tarlac 0.227 68 Low 67
Zambales 0.279 40 Moderate 39
Aurora 0.279 39 Moderate 51
CALABARZON 0.275 9 Moderate 12
Batangas 0.316 21 Serious 26
Cavite 0.252 56 Low 59
Laguna 0.295 32 Serious 32
Quezon 0.255 54 Low 54
Rizal 0.257 53 Low 56
MIMAROPA 0.27 12 Moderate 8
Marinduque 0.329 17 Alarming 13
Occ. Mindoro 0.181 80 Low 76
Or. Mindoro 0.334 12 Alarming 10
Palawan 0.193 77 Low 74
Romblon 0.395 4 Alarming 3
Region V 0.293 6 Moderate 6
Albay 0.266 48 Moderate 50
Cam. Norte 0.345 8 Serious 9
Cam. Sur 0.273 43 Moderate 44
Catanduanes 0.212 70 Low 68
Masbate 0.332 13 Serious 15
Sorsogon 0.331 14 Serious 18
Region VI 0.279 8 Moderate 7
Aklan 0.202 72 Low 73
Antique 0.33 15 Serious 11
Capiz 0.299 29 Serious 29
Iloilo 0.259 52 Moderate 53
Negros Occ. 0.299 30 Serious 28
Guimaras 0.147 83 Low 83
Moderate
Low
Moderate
Low
Moderate
Low
Low
Serious
Serious
Low
Low
Moderate
Low
Alarming
Moderate
Low
Low
Serious
Low
Serious
Low
Low
Low
Low
Low
Low
Low
Moderate
Low
Low
Low
Low
Low
Low
Low
Low
Low
–
Low
Low
Low
Low
Low
Low
Severity
Low
Low
Region VII 0.305 4 Serious 4
Bohol 0.323 19 Serious 19
Cebu 0.288 34 Moderate 36
Negros Or. 0.33 16 Serious 16
Siquijor 0.371 7 Alarming 7
Region VIII 0.305 3 Serious 3
E. Samar 0.249 59 Low 58
Leyte 0.315 22 Serious 23
N. Samar 0.411 2 Alarming 2
W. Samar 0.284 35 Moderate 35
So. Leyte 0.201 74 Low 71
Biliran 0.339 10 Serious 8
Region IX 0.318 2 Serious 2
Zambo del Norte 0.306 25 Serious 20
Zambo del Sur 0.314 23 Serious 24
Zambo Sibugay 0.301 27 Serious 27
Isabela City* – – –
Region X 0.281 7 Moderate 9
Bukidnon 0.237 64 Low 65
Camiguin 0.187 78 Low 80
Lanao del Norte 0.293 33 Serious 34
Misamis Occ. 0.381 5 Alarming 5
Misamis Or. 0.272 44 Moderate 43
Region XI 0.236 15 Low 15
Davao del Norte 0.202 73 Low 72
Davao del Sur 0.242 62 Low 61
Davao Oriental 0.328 18 Serious 17
Compostela 0.194 76 Low 79
Region XII 0.299 5 Moderate 5
N. Cotabato 0.31 24 Serious 25
S. Cotabato 0.265 50 Moderate 48
Sultan Kudarat 0.334 11 Alarming 14
Sarangani 0.275 42 Moderate 41
Cotabato City 0.439 1 Alarming 1
NCR 0.233 16 Low 16
District 1 0.238 63 Low 64
District 2 0.204 71 Low 75
District 3 0.248 60 Low 62
District 4 0.268 46 Moderate 45
CAR 0.222 17 Low 17
Abra 0.3 28 Serious 30
Benguet 0.2 75 Low 77
Ifugao 0.152 82 Low 82
Kalinga 0.317 20 Serious 22
Mt. Province 0.22 69 Low 69
Apayao 0.184 79 Low 78
ARMM 0.323 1 Serious 1
Basilan 0.261 51 Moderate 46
Lanao del Sur 0.252 55 Low 52
Maguindanao 0.305 26 Serious 21
Sulu 0.395 3 Alarming 4
Tawi-tawi 0.381 6 Alarming 6
Caraga 0.271 11 Moderate 11
Agusan del Norte 0.277 41 Moderate 40
Agusan del Sur 0.235 66 Low 63
Surigao del Norte 0.345 9 Alarming 12
Surigao del Sur 0.242 61 Low 55
Note: * incomplete data
Category Values
Low 0.000 – 0.256
Moderate 0.257 – 0.291
Serious 0.292 – 0.326
Alarming 0.327 – 1.000
Source: The Development of a Hunger Index for the Philippines: Methodology and
Results, LGBersales,2007
Low
Low
Low
Low
Serious
Low
Moderate
Alarming
Serious
Low
Low
Moderate
Low
Moderate
Low
Low
Moderate
Low
Low
Low
Low
Low
Alarming
Low
Low
Moderate
Moderate
Low
Moderate
Low
Low
Moderate
Low
Low
Serious
Low
Low
–
Low
Low
Low
Moderate
Moderate
Moderate
Moderate
Alarming
Low
Low
Serious
Serious
Moderate
Low
Moderate
Moderate
Moderate
Low
Moderate
Table 4. Hunger Index by Conponent Indicators by Region and Province, 2003
Region Province
Proportion of
HH with less
100% energy
adequacy
Proportion of
nderweight
children under
5 years
U5 MR
(expressed in
proportion)
Hunger Index
(equal
weights)
Hunger Index
(unequal
weights)
PHILIPPINES 0.569 0.2077 0.04 0.2722 0.235
Region XII Cotabato City 0.8 0.5001 0.0171 0.4391 0.383
Region VIII N. Samar 0.647 0.4075 0.178 0.4108 0.377
MIMAROPA Romblon 0.765 0.3337 0.0867 0.3951 0.347
ARMM Sulu 0.863 0.2583 0.0648 0.3953 0.339
Region X Misamis Occ. 0.818 0.2562 0.0698 0.3814 0.329
ARMM Tawi-tawi 0.875 0.2223 0.0448 0.3807 0.323
Region VII Siquijor 0.75 0.3333 0.0306 0.3713 0.32
Region VIII Biliran 0.625 0.3332 0.0588 0.339 0.299
Region V Cam. Norte 0.75 0.2223 0.0635 0.3453 0.297
MIMAROPA Or. Mindoro 0.6 0.333 0.0688 0.334 0.296
Region VI Antique 0.538 0.4002 0.0531 0.3304 0.296
Caraga Surigao del Norte 0.808 0.1689 0.0579 0.3449 0.293
MIMAROPA Marinduque 0.616 0.2859 0.0863 0.3294 0.292
Region XII Sultan Kudarat 0.657 0.2979 0.0471 0.334 0.291
Region V Masbate 0.646 0.3044 0.0444 0.3316 0.289
Region VII Negros Or. 0.665 0.2898 0.0349 0.3299 0.285
Region XI Davao Oriental 0.65 0.2941 0.041 0.3284 0.285
Region V Sorsogon 0.708 0.2389 0.0458 0.3309 0.284
Region VII Bohol 0.583 0.3429 0.042 0.3226 0.284
Region IX Zambo del Norte 0.542 0.3334 0.0428 0.3061 0.27
ARMM Maguindanao 0.565 0.2882 0.0603 0.3045 0.269
CAR Kalinga 0.692 0.2503 0.0079 0.3167 0.268
Region VIII Leyte 0.718 0.1818 0.0456 0.3151 0.268
Region IX Zambo del Sur 0.703 0.193 0.0458 0.3139 0.268
Region XII N. Cotabato 0.623 0.2876 0.0194 0.31 0.267
CALABARZON Batangas 0.724 0.2003 0.0238 0.316 0.267
Region IX Zambo Sibugay 0.564 0.3078 0.0327 0.3015 0.264
Region VI Negros Occ. 0.557 0.2962 0.0431 0.2988 0.262
Region VI Capiz 0.625 0.2165 0.0551 0.2989 0.259
CAR Abra 0.643 0.214 0.0443 0.3005 0.258
Region I La Union 0.578 0.2679 0.0398 0.2952 0.257
CALABARZON Laguna 0.67 0.1863 0.0291 0.2951 0.25
Region II Cagayan 0.473 0.3271 0.0423 0.2808 0.25
Region X Lanao del Norte 0.696 0.1255 0.0577 0.2931 0.249
Region VIII W. Samar 0.583 0.2342 0.0336 0.2836 0.245
Region VII Cebu 0.654 0.1698 0.0392 0.2877 0.245
Region II N. Vizcaya 0.562 0.2542 0.03 0.2821 0.244
Region I Ilocos Sur 0.583 0.2351 0.0308 0.283 0.244
Region III Zambales 0.529 0.2728 0.0346 0.2788 0.244
Caraga Agusan del Norte 0.548 0.2182 0.065 0.2771 0.243
Region XII Sarangani 0.5 0.2964 0.0277 0.2747 0.241
Region I Ilocos Norte 0.458 0.3073 0.0416 0.269 0.239
Region X Misamis Or. 0.583 0.1791 0.0536 0.2719 0.235
Region V Cam. Sur 0.59 0.1922 0.0359 0.2727 0.234
NCR District 4 0.536 0.2362 0.0304 0.2675 0.232
ARMM Basilan 0.572 0.0624 0.1484 0.2609 0.232
Region I Pangasinan 0.521 0.2343 0.0401 0.2651 0.231
Region XII S. Cotabato 0.564 0.1763 0.0548 0.265 0.229
Region III Pampanga 0.587 0.1974 0.0173 0.2672 0.227
Region V Albay 0.605 0.1511 0.0411 0.2658 0.226
Region III Aurora 0.8 0 0.0365 0.2788 0.226
ARMM Lanao del Sur 0.428 0.2797 0.0478 0.2518 0.225
Region VI Iloilo 0.561 0.1581 0.0582 0.2591 0.224
CALABARZON Quezon 0.5 0.2273 0.0385 0.2553 0.223
Caraga Surigao del Sur 0.353 0.3216 0.0512 0.2419 0.22
CALABARZON Rizal 0.572 0.1689 0.0315 0.2575 0.22
Region II Isabela 0.486 0.2343 0.0315 0.2506 0.218
Region VIII E. Samar 0.471 0.263 0.0126 0.2489 0.216
CALABARZON Cavite 0.554 0.1673 0.0339 0.2517 0.215
Region III N. Ecija 0.569 0.1552 0.03 0.2514 0.214
Region XI Davao del Sur 0.472 0.2039 0.0486 0.2415 0.212
NCR District 3 0.551 0.1605 0.0318 0.2478 0.211
Caraga Agusan del Sur 0.408 0.2769 0.0202 0.235 0.207
NCR District 1 0.515 0.1589 0.0386 0.2375 0.204
Region X Bukidnon 0.513 0.1729 0.0264 0.2374 0.203
Region III Bulacan 0.574 0.0898 0.0428 0.2355 0.198
Region III Tarlac 0.46 0.1883 0.0334 0.2272 0.197
Region V Catanduanes 0.294 0.2858 0.0555 0.2118 0.194
CAR Mt. Province 0.453 0.167 0.0407 0.2202 0.191
Region II Quirino 0.666 0 0.0262 0.2307 0.186
Region VIII So. Leyte 0.399 0.148 0.0568 0.2013 0.177
Region XI Davao del Norte 0.406 0.147 0.0527 0.2019 0.177
Region VI Aklan 0.461 0.1001 0.0451 0.2021 0.173
MIMAROPA Palawan 0.332 0.1931 0.0531 0.1927 0.173
NCR District 2 0.518 0.0658 0.0269 0.2036 0.169
MIMAROPA Occ. Mindoro 0.25 0.2272 0.0652 0.1808 0.167
CAR Benguet 0.527 0.0392 0.0327 0.1997 0.165
CAR Apayao 0.333 0.1821 0.0363 0.1838 0.163
Region XI Compostela 0.5 0.0466 0.036 0.1942 0.162
Region X Camiguin 0.5 0 0.0619 0.1873 0.157
Region III Bataan 0.4 0.1053 0.0278 0.1777 0.152
CAR Ifugao 0.308 0.1051 0.0414 0.1515 0.133
Region VI Guimaras 0.4 0 0.0397 0.1466 0.122
Region IX Isabela City* 0.8 0.5 – – –
Region II Batanes* – – 0.0785 – –
Source: The Development of a Hunger Index for the Philippines: Methodology and Results, LGBersales,2007