Indonesia is an archipelago state located in South East Asia between the Indian and Pacific Oceans. It has more than 17,000 islands with 6,000 of those for good inhabited. The entire land country is about 1.9 million square kilometers and in 2008, it had a population of about 240 million. Administratively, the Republic of Indonesia is divided into 33 states. The World Bank ( 2009 ) has classified it as a lower in-between income state with per capita GDP of PPP amounting to US $ 3,979 in 2008. Harmonizing to the United Nation Development Programme ( UNDP, 2009 ) , the Human Development Index ( HDI ) for Indonesia in 2007 is at 0.734 and its ranking is 111. The official figures of some indexs related to poorness and inequality steps from 1996 to 2008 are shown in Table 1. The tendency of inequality is by and large increasing while the tendency of the poorness index is diminishing, except for the period affected by the crisis in 1999, where inequality decreases and the head count index additions.
Table 1 Trend in Poverty and Inequality Related Indicators 1996-2008
Beginning: BPS Statistics Indonesia ( 1996, 2003, 2007, 2008 )
Susenas Expenditure Data
In the present survey, we approach the income distribution utilizing the family outgo informations obtained from the National Socio-Economic Survey ( Susenas ) . Susenas is a cross-sectional family study for Indonesia which provides national coverage and is available over an extended clip period. A portion of Susenas is conducted yearly roll uping information on the features of over 200,000 families and over 800,000 persons. This portion of Susenas is known as the nucleus Susenas. Another portion is conducted every 3 old ages, roll uping information on really elaborate ingestion outgos on nutrient and non-food points from about 65,000 families. This portion is popularly known as the ingestion faculty Susenas.
The dataset is created by unifying the nucleus and the faculty for 1996, 1999, 2002, 2005 and 2008. The created dataset has a combination of information on family ingestion from the ingestion faculty Susenas and family features from the nucleus Susenas. The analysed variable is the monthly family outgo of nutrient and non-food ingestion. Naturally, the outgo degree may change harmonizing to comparative monetary values, demographic factors and penchants. In fact, the features of urban and rural countries are really different in footings of these facets, so this environment will really find the wellbeing reading in Indonesia. For this ground, we analyse urban and rural countries individually.
The entire sample size is about 60,000 each twelvemonth. The fluctuation in different study old ages is from assorted restrictions of the informations. For illustration, in 2002 due to some political instability, the study did non cover 4 states, cut downing the sample size at the national degree. The intervention of some losing and utmost values in the meeting of the Susenas nucleus and faculty datasets has besides reduced the sum of informations being processed.
Monetary value Adjustment
In order to do the outgo informations set comparable across different study old ages, the informations were corrected for rising prices utilizing the consumer monetary value index ( CPI ) . The CPI index for nutrient and non-food groups reported by Badan Pusat Statistik ( BPS ) Statistics Indonesia was constructed for urban monetary values collected from 27 metropoliss in 1996, 44 metropoliss in 1999 and 45 metropoliss in 2005 and 2008. Due to a restriction in informations handiness, we used urban monetary value indices as placeholders for the alterations in monetary values for the rural countries in each state. For urban and rural countries that were non covered in the CPI series, we approximate them by utilizing the CPI values of the neighbouring metropoliss. In such vicinities, we expect to hold rather similar features in footings of the monetary value index. In this paper, family outgos were adjusted to existent outgo at 2002 monetary values.
Family outgo besides has to be adjusted for the demographic differences to integrate adult-child fluctuation in family composing every bit good as positive economic systems of graduated table as family size additions. Therefore, alternatively of utilizing per capita, we deflated the family outgo by utilizing an equality graduated table which accommodates an grownup tantamount graduated table and economic systems of graduated table. Equivalence graduated table pattern has been researched extensively in the literature. As puting the equality graduated table was non our chief research inquiry, we have opted to utilize the preparation used by Banks and Johnson ( 1994 ) and Jenkins and Cowell ( 1994 ) , which is recommended by Deaton and Zaidi ( 2002 ) for the instance of developing states such as Indonesia. It is given by
( 3.1 )
where myocardial infarction is the figure of big equivalents in family I while sodium, I and North Carolina, one correspondingly denote the figure of grownups and kids in family I. The parametric quantity i?¦ is the cost of a kid relation to that of an grownup while the parametric quantity i?¬ represents economic systems of graduated table in the cost of tantamount grownups. In this survey, grownups were defined as family members aged 15 old ages and over because 15 is regarded as the age for get downing work. Further disaggregation of age groups and gender were non considered as demographic factors.
For the instance of hapless economic systems, Deaton and Zaidi ( 2002 ) suggest puting i?¦ every bit low as 0.3 and i?¬ near to 1. The recommended figures are motivated by the fact that kid costs in hapless states are comparatively cheap, and families in hapless states devote a larger portion of their outgo to nutrient. So, there would non be much infinite for economic systems of graduated table. However, these figures may non be applicable for the instance of Indonesia as the outgo portions on nutrient have declined over the old ages, from about 0.65 in 1996 to around 0.50 in 2008. The displacement in expenditure portions is besides likely to impact the magnitude of kid costs.
When seeking for the appropriate values for the economic systems of graduated table, i?¬ , and the size of kids comparative to grownups, i?¦ , we started by repairing the bounds for the equality graduated tables based on a recent survey by Lancaster and Ray ( 2002 ) . The values of i?¦ and i?¬ were so verified utilizing a simple generalisation of the Engel methodological analysis developed by Valenzuela ( 1996 ) . In drumhead, we have arrived at the decision that i?¦ = 0.85 and i?¬ = 0.8 are realistic values of the kid cost and economic systems of graduated table for Indonesia. The features of the size adjusted outgos can be seen in Table 2 while the associated histograms are reported in Figures 3.1-3.5.
The mean, average and standard divergence in 1999 decreased to some extent compared to the 1996 figures due to the impact of the Asian pecuniary crisis. Those statistics in general increased bit by bit after 1999, except for the urban countries in 2008 where the figures decreased somewhat once more. The urban countries were besides found to hold mean, average and standard divergence about twice every bit high as those in the rural countries. The histograms by and large show a typical uni-modal and right-skewness form for income distributions, with urban countries seemingly holding a more skewed and dispersed distribution than the rural countries.