NOTES

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. An earlier helpful survey is Meyer (2002). This draws out some of the methodological

problems in assessing impact, and surveys a number of important studies available at

the time of writing (around 2001). Morduch (1999) is an extremely authoritative earlier

survey focusing on both conceptual and empirical questions.

2. Lapenu and Zeller (2001: 28) Table 16.

3. Morduch (2003) points out that, although this argument may be true, the data in Hulme

and Mosley’s book cannot be used to infer this since the arithmetic basis for their

comparison of income changes for different categories of borrowers biases their results

in favor of their conclusion.

4. An important attempt to address this problem has been the Income Generation for

Vulnerable Group Development (IGVGD) program run by BRAC in Bangladesh, which

combines measures of livelihood protection (food aid) with measures of livelihood

promotion (skills training and micro-credit). Hence micro-credit is provided as part of

a package approach. Matin and Hulme (2003) survey the evidence on how far the benefi ts

of this program actually reach the core poor and conclude that although the program

was more successful than more conventional micro-credit schemes, none the less many

target households were still missed.

5. Coleman (2001) has a useful non-technical explanation of the diffi culties of applying

this approach and eliminating ‘selection’ and ‘placement’ bias in micro-credit studies.

6. See Hulme (1999) for a discussion of different approaches to impact. He points out that

despite their cost in funds and time, such rigorous studies involving detailed sample

surveys are the most common approach where the aim is to establish impact for policy

or investment purposes.

7. Technically the study is rigorous in employing a two-stage instrumental variable approach

along with a household fi xed-effects method to control for possible endogeneity bias,

particularly the fixed unobserved characteristics of households (that is the more

entrepreneurial amongst the poor are those who borrow, and these may do better

anyway).

8. Poverty is based on a calorie intake of 2112 and extreme poverty on one of 1739.

9. This debate, which in part centers around details of econometric estimation, has not

been resolved. An unpublished paper by Pitt reworks the original analysis to address

the concerns of Morduch and is said to confi rm the original results (Khandker, 2003,

footnote 1).

10. Unlike the Khandker studies this data picks up households before they joined a microcredit

scheme. Their vulnerability measure is broader than simply fluctuations in

consumption.

11. Patten et al. (2001) fi nd evidence that the micro-fi nance side of the Indonesian banking

system performed much more robustly during the macro crises of the late 1990s than

did the commercial banking sector.

12. Fujita (2000) makes this point in the context of Bangladesh.

13. The study on this scheme by Wodon (1998) appears considerably more sophisticated than

the other studies and compares costs with the future stream of estimated benefi ts to the

poor in terms of gains from education. The ratio for this activity may not be directly

comparable with the other fi gures in the table.

14. It should be noted that the benefi ts from Grameen lending found in Khandker (2003),

which are almost half of those found in his earlier study, imply considerably higher costeffectiveness

ratios than those reported in Table 7.3, unless there has been a corresponding

rise in the effi ciency of operations.