IMPACT ON COMMERCIALIZATION 35
Table 3.4 Distribution of households across treatment and comparison kebeles
Household type
Cooperative nonmembers Cooperative members Total
Comparison kebeles
1,702 0
1,702
Treatment kebeles
680 150 830
Total 2,382 2,532
150
Source: Authors’ calculations based on data from ESCS (2005). Note: A kebele is a peasant association, the smallest administrative unit in Ethiopia.
is the household’s membership status. Domain dummies are used to ensure matching within the domains. Household characteristics include measures of the household’s assets (such as education level, radio ownership, nonfarm income, landholding, and livestock) that are introduced linearly as well as quadratically to augment the model’s predictive power. Finally, a set of dummy variables is included to account for the household’s cultivation of particular cereal crops.4
We must also consider that a household’s production of a particular cereal may be in response to participation in the cooperative. The estimated impact would then be downward biased, as it might not take into account a house- hold’s switch into production of higher profit crops. However, the focus of the present monograph is the cooperatives’ impact on smallholders’ market- ing behavior. As such, one wants to compare marketing behavior of house- holds engaged in similar production, regardless of whether this was driven by the cooperative. In addition, the production of particular cereals is largely driven by soil and weather conditions in Ethiopia: teff is mainly cultivated in highland areas north of Addis Ababa, maize in the lowlands south of Addis Ababa, sorghum in the northwest and east, and barley along a north-south meridian in the middle of the country (CSA, EDRI, and IFPRI 2006). The probit estimation is better identified when undertaken on treatment
kebeles only, where the choice to join a cooperative exists (see Gotland et al. 2004 for a discussion). We report estimates of the coefficients in Table 3.5. We also report the associated p-values, although the purpose here is not to identify particular relationships but rather to maximize the predictive power of the model. Such an approach relies on out-of-sample prediction to gener- ate a propensity score for the comparison households, however. To assess
4 All households in this sample are involved in cereal production.
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92 |
Page 93 |
Page 94 |
Page 95 |
Page 96 |
Page 97 |
Page 98 |
Page 99 |
Page 100 |
Page 101 |
Page 102 |
Page 103