food inflation, job misery, and hunger incidence in the ......food inflation, job misery, and hunger...
TRANSCRIPT
Food Inflation, Job Misery, and Hunger Incidence in the Philippines:
A Panel Data Analysis
Reniel Louise Bagsit, Melrose Ivy Mendoza,
Maria Lorena Tabao and Dennis S. Mapa
School of Statistics
University of the Philippines Diliman
08 March 2019
Social Weather Stations (SWS)
Motivation of the Research
Food prices spiked in 2018, after a period of relatively low and stable price
regime in 2015 to 2017.
Data from the Philippine Statistics Authority (PSA) shows that the average
food inflation rate in 2018 was 6.9 percent, compared to just 3 percent in
2017.
The increase in food inflation was mainly due to the increasing prices of rice
and fish.
Higher food prices hurt the poor more since a large percentage of their
expenditures is allotted to food.
This paper looks at the impact of food inflation and job misery index,
defined as the sum of unemployment and underemployment rates, on
hunger incidence in the country using panel data.
The paper analyzed the impact of food prices and job misery on self-rated
hunger using the sub-national panel of the SWS, namely: National
Capital Region, Balanced Luzon, Visayas and Mindanao.
The paper employs the dynamic panel models suggested by Arellano and
Bond.
Objectives of the Research
Mapa, Han and Estrada (2011) and Mapa, Castillo and Francisco (2015)
showed that spikes in Food Inflation, particularly in the price of Rice,
increased Self-Rated Hunger incidence in the succeeding quarters after
the spikes in prices.
Reyes, Celia M., Alellie B. Sobrevinas, Joel Bancolita and Jeremy de
Jesus (2009), looked at the Impact of Changes in the Prices of Rice and
Fuel on Poverty.
Son, Hyun H. (2008) showed that inflation hurt the poor using regional
analysis
Previous Studies on Inflation-Poverty Nexus
Two Inflation Rates: Headline and the Poorest 30% of the HHs
-2
0
2
4
6
8
10
2013 2014 2015 2016 2017 2018
HEADLINE POOREST30
Figure 1. Headline Inflation (Base 2012) and Inflation of the Poorest 30% (Base 2000)
January 2013 to January 2019 (Source: PSA)
Weight in the CPI Basket (Base Year 2012)
Commodity Groups Poorest 30% of HHs Ave Household (Headline)
Food and Non-Alcoholic Beverages 60.89 38.34
Housing, Water, Electricity, Gas and other Fuels 18.00 22.04
Transport 4.21 8.07
Health 2.03 3.89
Clothing 2.07 2.93
Alcoholic Beverages and Tobacco 2.48 1.58
Weights of the Commodity Groups for the CPI Basket of the Average (Headline)
and Poorest 30% of Households
Authors’ Computation using PSA data (Rice - 22 percentage points; Fish - 11 percentage
point; Vegetables - 8 percentage points; and Meat - 6 percentage points).
3.4
3.8
4.34.5 4.6
5.2
5.7
6.46.7 6.7
6.0
5.1 5.2
4.0
4.4
5.2 5.3 5.3
5.9
6.7
7.6
8.2 8.1
7.1
6.0 6.1
JAN FEB MAR APR MAY JUNE JULY AUG SEPT OCT NOV DEC AVERAGE
Headline Poorest 30%
Headline Inflation Rate and Inflation Rate of the Poorest 30% of the Households
(January to December 2018; Base Year 2012)
2.6
3.2 3.2 3.4
3.2
2.8 2.6
2.8
3.3 3.3 3.1
3.3 3.1
4.0
4.4
5.2 5.3 5.3
5.9
6.7
7.6
8.2 8.1
7.1
6.0 6.1
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC AVERAGE
2017 2018
Inflation Rate of the Poorest 30% of the Households in 2017 and 2018
(Base Year 2012)
Commodity Groups Percentage Point Increase Contribution
Food and Non-Alcoholic Beverages 2.31 77%
Tobacco and Alcoholic Beverages 0.32 11%
Housing, Water, Electricity, Gas and Fuels
(HWEGF) 0.22 7%
Transport 0.07 2%
Sources and Contribution of Increase in Inflation Rate of the Poor Households in 2018
High Inflation Rate among POOR Households:
Critical Constraints to Poverty Reduction Efforts
High income growth and low inflation are key to poverty reduction effort.
In the previous Family Income and Expenditure Survey (FIES) years, significant
reduction in poverty was achieved only during the period of high income growth with
low inflation among the POOR households.
2006 2009 2012 2015
Per Capita Poverty Threshold (National) 13357 16871 18935 21753
Average Annual Growth Rate
in Per Capita PT (in %) Base Year 8.10 3.92 4.73
Increases in Per Capita Poverty Threshold from FIES
(2006, 2009, 2012, 2015)
Income Growth and Inflation Rate play Critical Roles in Poverty Reduction
Significant
reduction
in poverty
from 2012
to 2015
No
significant
reduction
No
significant
reduction
6.0 - 6.4 %
Glimpse on the 2018 Official Poverty Count?
Self-Rated Poverty inched up
significantly to 52 percent
during the 3rd quarter of 2018
from 42 percent in the 1st quarter
2018.
Two successive quarters (2nd and
3rd quarters 2018) increase.
Total of 10 percentage points
increase.40
45
50
55
60
65
70
2000 2002 2004 2006 2008 2010 2012 2014 2016 2018
Poverty Incidence
Poverty Trend (HP Filter)
Self Rated Poverty Incidence from 1st Quarter 2000 to 3rd Quarter 2018
(Source: SWS)
4
8
12
16
20
24
2000 2002 2004 2006 2008 2010 2012 2014 2016 2018
HUNGER TREND_HUNGER
Self-Rated Hunger Incidence from 1st Quarter 2000 to 3rd Quarter 2018
(Source: SWS)
Self-Rated Hunger Incidence
increased significantly during
the 3rd Quarter of 2018 to 13.3
percent, from 9.4 percent during
the 2nd Quarter of 2018
3.9 percentage points increase.
Food Inflation explains Hunger Incidence
Methodology (Arellano-Bond)
This paper examines the dynamic patterns of hunger incidence and the effects of the
determinants of hunger, food inflation and job misery index, using the quarterly sub-
national time series data from SWS survey, covering the period from 1st quarter of 2012
to 3rd quarter of 2018, using dynamic panel data.
There are several advantages of using panel data, for one panel data increases the
precision in the estimates by pooling the data and more importantly, there is a possibility
of consistent estimation of fixed effects model which allows for the unobserved
heterogeneity to be correlated with the regressors.
Unobserved effects lead to the problem of omitted variable which can make our
estimated coefficients biased and inconsistent when using the least squares (LS)
regression.
Methodology (Arellano-Bond)
The generic dynamic panel model is given by,
Hit is the Self-Rated Hunger Incidence for the sub-national group i at quarter t, there are
4 sub-national groupings representing the National Capital Region (NCR), Visayas
Islands Group, Mindanao Islands Group and the Luzon Group;
FIit is the Food Inflation in the sub-national group i at quarter t;
JMit is the Job Misery Index defined as the sum of the employment rate and the
unemployment rate;
ui is the unobserved effects and εit is the error term.
itiitittiit uJMFIHH 221,10
Methodology (Arellano-Bond)
The problem with the unobservable random variable, 𝑢𝑖 is that if the 𝐶𝑜𝑣(𝑥𝑗
, 𝑢) ≠ 0(i.e. the explanatory variables 𝑥𝑗 are correlated with u for some j), and u is included in
the error term, then it will lead to the problem of endogeneity if not treated correctly.
This problem implies that our estimated 𝛽 coefficients will be biased and inconsistent.
The model is estimated using the Arellano-Bond procedure.
Variable Estimated Coefficient Robust SE P-value
Food Inflation (in %) 0.6766 0.1413 0.0000
Job Misery Index (in %) 0.7646 0.1821 0.0000
Constant -10.4382 4.6520 0.0250
Total Hunger
Lag 1 0.2570 0.0611 0.0000
Lag 2 0.2146 0.0797 0.0070
Determinants of Self-Rated Hunger Incidence using Panel Data (NCR, Luzon, Visayas and Mindanao)
(Dynamic Panel Data Analysis; Arellano-Bond Model)
Econometric Model for Self-Rated Hunger Incidence
Dynamic Panel Model for NCR, Luzon, Visayas and Mindanao
(2012Q1 to 2018Q2)
Job Misery Index is the sum of the unemployment and underemployment rates
Econometric Model for Self-Rated Hunger Incidence
Dynamic Panel Model for NCR, Luzon, Visayas and Mindanao
(2012Q1 to 2018Q2)
A one-percentage point increase in Food Inflation leads to an increase in the average
Self-Rated Hunger Incidence by about 0.68 percentage point.
A one-percentage point increase in the Job Misery Index leads to an increase in the
average Self-Rated Hunger Incidence by about 0.76 percentage point.
3rd Quarter 2018 3rd Quarter 2017 Increase/Decrease
Food Inflation (in %) 8.4 3.0 5.4
Job Misery Index (in %) 22.6 21.9 0.7
Contribution of Food Inflation on Hunger Incidence 3.7 percentage points
Contribution of Job Misery Index on Hunger Incidence 0.5 percentage point
Near Poor Household Classification (Data from 2015 FIES)
Percentage 1 to 1.1 times of PT 1 to 1.2 times of PT 1 to 1.3 times of PT
Households 4.21 8.45 12.25
Population 4.94 9.76 13.92
Near Poor Households
A NEAR-POOR household is a NON-POOR household at a given time (FIES Year), but
with higher probability of becoming POOR in the future (example, in 3 years), due to
“shocks” affecting the household such as health, economic, disaster and the like. In the
literature, also referred to as TRANSIENT POOR.
Vulnerable Sectors affected by Inflation
Source of Info-graphics: Philippine Statistics Authority (PSA)
Group Average Per Capita Income of Near Poor
(10 percent above PT 2015)
Near Poor (Overall) 22,748.00
Percent Higher than PT 5.00
Fishers (Near Poor) 21,845.00
Percent Higher than PT 0.40
Farmers (Near Poor) 22,392.00
Percent Higher than PT 3.00
Women (Near Poor) 22,778.00
Percent Higher than PT 5.00
Seniors (Near Poor) 22,693.00
Percent Higher than PT 4.00
Near Poor Households (Basic Sectors)
Near Poor Population (Basic Sectors; 2015 FIES)
Fishers Farmers Women Seniors
Percent of Population 8.07 6.36 5.28 4.23
Mostly Found in Regions 5, 11, ARMM, 8
and CARAGA
Regions ARMM, 10, 8, 9
and CARAGA
Regions ARMM, 8,
CARAGA, 12, 10 and 5
Regions ARMM,
CARAGA, 8, 12, 10
Near Poor Households (Basic Sectors)
Conclusions
High inflation rates experienced by the poor households is a threat to any
poverty reduction effort.
Using official data on income and poverty from the PSA over the last ten
(10) years, the analysis showed that significant reduction in poverty was
achieved only during period of high income growth among the poor
households, coupled with relatively low inflation rate experienced by the
poor.
Spikes in self-rated hunger and self-rated poverty incidence in 2018 may be
indicators on what to expect on the official poverty report.
Thank you and good afternoon.
Reniel Louise Bagsit, Melrose Ivy Mendoza,
Maria Lorena Tabao and Dennis S. Mapa
School of Statistics
University of the Philippines Diliman
08 March 2019
Social Weather Stations (SWS)