IMPACT ON COMMERCIALIZATION 43
of the mean price, which in turn may severely affect the precision of our estimates. To test the importance of such potential bias, we compute the average treatment effects when only the Oromia region is considered in the analysis. (As shown in Table 3.2, Oromia is the only region where a suf- ficiently large sample of treatment and comparison kebeles exist.) Results reported in panel B-1, although less precise, are qualitatively similar to those of panel A.
Second, we investigate whether the inclusion of NGO-created coopera- tives in the final sample may affect the results. Indeed, if NGOs deliberately choose the kebeles in which they intervene based on criteria linked to the likely performance of the organizations they help set up, results may be artificially high. In panel B-2, we present estimates from a sample excluding kebeles where NGO-created cooperatives exist. The estimates are reduced in their magnitude and precision, although the main features remain. Third, we check whether our use of the sample of kebeles with coopera- tives only, to derive the propensity score estimates, may have affected the results. Results are reported in panel B-3; they do not show any meaningful differences from those of panel A.
Next we investigate how the estimates may be affected by more represen- tative but arguably more biased samples. A first concern may come from our limitation to cereal crop cooperatives, when evidence suggests that coopera- tives are often more efficient when operating with high-value products, such as coffee. As discussed in Chapters 1 and 2, cereals do, however, constitute the most important crops for smallholders’ livelihood in many Sub-Saharan African countries, particularly in Ethiopia.
Further, in panel C-1, we use a sample containing all regions, including Tigray, Beneshangul-Gumuz, and Harari. Again, the results do not clearly differ from those of panel A, although the impact of cooperatives on output prices seems greater in magnitude than for the fully restricted sample. The same conclusions are drawn from panels C-2 and C-3, where the sample includes all kebeles with member-created cooperatives, first restricted to only three regions (Amhara, Oromia, and SNNP) and then extended to cover all regions. Finally, in panel C-4, results are based on the sample of panel A but are further restricted to only kebeles where an externally created cooperative exists. The comparison is then done between cooperative members and nonmembers living in the same kebeles. Again, the results do not change in nature.
Another robustness check uses parametric estimations that control for several household- and community-level characteristics (Table 3.8). Because participation in a cooperative may directly affect the production levels of
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