Forest dependence and income inequality in rural Ethiopia: evidence from Chilimo-Gaji community forest users

Abstract

The aim of the study was to examine the determinants of forest dependence and the role of community forest on income inequality in rural Ethiopia. Regression results, using Heckman’s two-stage estimation method, suggest that the probability of households’ participation in low-return forest activities is determined by farm size, number of male members in the household and distance from the forest plot to the household’s homestead. Further, the likelihood of households’ participation in high-return forest activities is determined by the number of male household members, the distance from the households’ homestead to the community forest block and being a member of the forest user group’s executive committee. Using instrumental variable method, we found that, in relative terms, households with more non-forest income are less likely to depend on forest commons for their livelihood. Further, wealthier households are less dependent on forest products for their livelihood. We also found that forest products play a crucial role in reducing income inequality in the study area. Income inequality increases by 24% when we exclude forest income from the calculation of inequality measure (Gini coefficient).

Keywords:

• forest dependence
• forest income
• inequality
• Ethiopia
• Africa

Acknowledgements

We feel greatly indebted to the financial support obtained from DAAD (Deutscher Akademischer Austausch Dienst) for the study.

Notes

1. Regarding the exact size of the forest area in the study site, one may realise that data on the size of the forest area vary from report to report. For example, the forest area as of 2011 was 4944 ha according to the documents obtained from Chilimo-Gaji community forest users’ union. On the other hand, according to Kebebew (2012) (FARM-Africa), the forest area is 5627.7 ha. Further, according to Bekele and Tsegaye (2005), the forest cover in the study area is estimated to be 5800 ha.

2. See Section 4.3 for the specific activities included in LRFA and HRFA.

3. A household in this case includes all members that share consumption of food and non-food items. Moreover, it includes workers and servants if they have stayed together in the same household at least for half a year.

4. It is an agrarian system that mixes crop production with the raising of livestock.

5. During the study time, 1 Euro = 24.56 ETB.

6. In the calculation of income, own consumption is computed at opportunity cost (i.e., how much the household could have spent, had the own consumption part was bought from the market) and included in the analysis with that understanding.

7. Note that in this case the sum of the participation rate is greater than 100% because there are some households that participate in different forest activities simultaneously.

8. From the group discussions, it was evident that farm size reflects the wealth status of a household in the study area.

9. We included dummy variables to account for villages (FUGs) but found no significant effect on forest dependence; hence, we have not included the village dummies in the report and can be obtained on request from the authors. To account for the possible existence of heteroscedasticity, we obtained robust standard errors, which are corrected for possible heteroscedasticity.