К оглавлению
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 
68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 
85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 
102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 
119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 
136 137 138 139 140 141 142 

1. The offi cial headcount poverty fi gure is published by the Central Bureau of Statistics

(Badan Pusat Statistik or BPS). Their poverty rate calculations are based on the annual

National Socio-Economic Survey (Survey Sosial Ekonomi Nasional or SUSENAS). The

full database consists of the core data based on over 200 000 households and several

module data, including a consumption module, which is repeated every three years

and covers about 65 000 households across Indonesia. We use 1996 as the pre-Crisis

benchmark because it was the last of BPS’s pre-Crisis socio-economic surveys. In the

1998 survey BPS changed its method for calculating poverty by revising the non-food

bundle. For example, the expenditure for schooling was revised to take account of the

nine years of compulsory schooling that had been introduced (previously it was only six

years). Then it revised the 1996 fi gure, adjusting to the 1998 methodology. In the table

we present both calculations of the 1996 fi gure.

2. Some observers have argued that the BPS calculation for 1998 was an overestimate,

and was not an accurate refl ection of the real situation. Several studies of post-Crisis

poverty measurements have produced a lower 1998 poverty rate than the BPS calculation;

for example, Frankenberg and Beegle (1999). Nevertheless, all agreed that poverty had

worsened considerably following the Crisis.

3. A SUSENAS-type survey but covering only around 10 000 households.

4. Excluding the troubled provinces of Aceh and Maluku.

5. Collected also by the BPS with a sample of 12 000 households in several rounds; May

1997, August 1998, December 1998, May 1999 and October 1999.

6. Gardiner (1999) used the SUSENAS core data to make independent poverty estimates

for February 1996, 1997 and 1998.

7. Frankenberg and Beegle (1999) used various versions of the Indonesian Family Life

Survey carried out by the RAND Institute and local universities.

8. Although the authors used the SUSENAS database, they applied a different technique

to make their calculations, so poverty rates for all sectors are different to the offi cial

fi gures. But here we are concerned more about the relative change between 1996–99 in

each sector.

9. By contrast, the transient poor are those whose consumption levels are somewhere near

or just below the poverty line. Transient poverty is generally the result of an economic

shock so that household income is inadequate to meet basic needs. Households in this

category still have the capacity to survive, and can improve their condition if the economy

strengthens and grows.

10. See also Dhanani and Islam (2002) for further discussion on ‘capability poverty’ with

regard to the post-Crisis situation in Indonesia.

11. This was shown in the results of a rapid survey of local respondents at the sub-district

(kecamatan) level conducted in October 1998 by BPS and the World Bank; see also

Feridhanusetyawan (2000:114–115). Eastern Indonesia, where headcount poverty levels

had always been very high, had also been seriously affected by the 1997–98 El Niñorelated


12. BKKBN conduct regular workshops to inform their workers about the procedures to be

followed during the registration process, and there is an evident sense of pride among

those who hold these positions. Typically, these village cadres who may include the wives

of prominent local people, schoolteachers and village offi cials have often been carrying

out the task over a number of years, and have the advantage of a close and intimate

knowledge of their local community.

13. The other levels, indicative of improving welfare status, were ‘prosperous families’ level 1,

level 2, level 3, and level 3 plus (keluarga sejahtera 1, keluarga sejahtera 2, keluarga

sejahtera 3, and keluarga sejahtera 3+); see Sumo and Soedjono (1995).

14. ALEK is alasan ekonomi (economic reasons).

15. See Daley and Fane (2002: 311) for further details on anti-poverty programs and the

amount of government budget expenditure that has been directed to such measures.

16. The second tranche of the World Bank Social Safety Net Adjustment Loan was eventually

canceled after fi eld verifi cation and monitoring surveys revealed evidence of gross excesses

and poor implementation in several of these programs.

17. The IDT program was run by the National Development Planning Board (BAPPENAS),

in coordination with the Ministry of Internal Affairs and the Central Bureau of Statistics


18. Using the 1996 exchange rate at Rp 2300 per US$.

19. The Podes survey is carried out by BPS approximately three times in every ten years,

usually immediately before one of the major censuses (population, agriculture, the

economy). The Podes questionnaires are completed by sub-district (kecamatan) offi cials

who rely on information collected from village offi cials in their area.

20. These variables were divided into three categories: social and economic characteristics,

housing and environment, and population. For a list of the variables used for the 1993

and 1994 selection process, see Alatas (1999: Appendix A).

21. For further details on the complexities of the BPS ranking and selection process for IDT,

see Alatas (1999: 4–6) and Sumarto et al. (1997: 12).

22. This report was part of an Asian Development Bank study, reviewing poverty

alleviation strategies within the agriculture sector for the Ministry of Agriculture and


23. The term padat karya had been used widely throughout Indonesia since at least the

early 1970s for small-scale village infrastructure activities that selected the required labor

entirely from within the village community. It probably also draws on both traditional

notions of how to conduct communal self-help activities and the corvée labor schemes

that were used by colonial and traditional rulers for public works projects.

24. See for example, Suryahadi et al. (1999: 12–14).

25. The program was mostly referred to in the press and by the general public simply by its

abbreviation: few outside the government bureaucracy ever really understood what the

full name of the program was or the precise nature of the benefi ts that it was intended

to deliver to benefi ciaries.

26. See Akhmadi et al. (1999). A brief English language summary of the central fi ndings

appears in SMERU (1999: 4–9).

27. The term OPK was adopted to distinguish this program from those market operations

(operasi pasar) that the National Logistics Agency conducted periodically by ‘dropping’

rice into the market place as a price stabilization measure. This had been carried out

frequently during 1997 in response to perceived shortages in the availability of rice and

other foodstuffs because of the El Niño drought.

28. For details of the disbursements of rice during each phase of the subsidized rice program

between 1998 and 2002, see Hastuti and Maxwell (2003: 8).

29. See Kusumastuti (1998: 18–19), who suggested a revision of the BKKBN eligibility

criteria and the application of some form of local decision-making as part of the targeting


30. For an assessment of this program based on a rapid fi eld assessment, see Hastuti et al.

(2000). The report includes a brief comparison between this program and the government’s

OPK program.

31. At the urging of several international agencies, in particular FAO and the World

Food Program, serious consideration was given to introducing a technique known as

the Vulnerability Analysis and Mapping (VAM) methodology. Using a multivariate

statistical approach and drawing on data from a range of sources (household consumption

and expenditure, nutrition and poverty levels, in combination with data on food crop

production levels), it was argued that this methodology could be used to create a series of

digital maps covering the entire country and coded to indicate levels of local vulnerability

to food insecurity in specifi c areas. Those areas at greatest risk of experiencing food

insecurity would be selected to receive the subsidized rice. The VAM methodology was

outlined and discussed at a Bulog workshop in Jakarta in November 2001, organized to

review the targeting issues.

32. See Hastuti and Maxwell (2003: 6–8) for a more detailed account of the determination

of quotas at national, province, district and local levels.

33. Jones and Hagul (2001: 217–218) also reveal that trends in enrollment rates have not

changed signifi cantly between males and females.

34. Pradhan and Sparrow, as quoted in Sparrow (2003b: 18), have calculated the monthly

household expenditure on education per student for primary, lower and upper secondary

schools in the 1997/98 academic year. These were Rp 9562, Rp 27 862 and Rp 53 243 (at

February 1999 prices) respectively. Average monthly education expenditure of the poorest

20 per cent of the population at these different levels was estimated to be Rp 4826, Rp

15 725 and Rp 31 549 per student in 1998. Hence, the scholarship provided a signifi cant

proportion of costs.

35. See the report by CIMU (2000a:12). CIMU stands for Central Independent Monitoring

Unit, a body established and funded jointly by several international agencies exclusively

to monitor the implementation of the SGP, as well as to investigate reported irregularities

in the implementation of the program.

36. This fi nding was also supported by Sparrow (2003b), while another study by Cameron

(2002) using a different data set, the 100 Village Survey, found that the scholarships

were received by 8.4, 13.6 and 9.6 per cent of primary, lower and upper secondary level

students, respectively.

37. As Sparrow (2003b: 12) argued, the 100 Village Survey contains a large number of

relatively poor villages. Thus the survey may not be representative of the actual conditions

in Indonesia.

38. The Asian Development Bank provided funding for 8 provinces, distributed in 4 phases

beginning in August 1998, while the State Budget funded 19 provinces in 2 phases

beginning October 1998 (Soelaksono et al., 2003: 5). Some of the State Budget allocations

were supported by the IMF loan package.

39. A system of community health centers, commonly known as Puskesmas (Pusat Kesehatan

Masyarakat), had been established during the previous decades, and had made basic

health care accessible to both urban and rural villages throughout Indonesia.