Importance, determinants and gender dimensions of forest income in eastern highlands of Ethiopia: The case of communities around Jelo Afromontane forest

Forest Policy and Economics 28 (2013) 1–7

Contents lists available at SciVerse ScienceDirect

Forest Policy and Economics journal homepage: www.elsevier.com/locate/forpol

Importance, determinants and gender dimensions of forest income in eastern highlands of Ethiopia: The case of communities around Jelo Afromontane forest Adanech Asfaw a, Mulugeta Lemenih b,⁎, Habtemariam Kassa c, Zeleke Ewnetu a a b c

Wondo Genet College of Forestry and Natural Resources, P. O. Bo 128, Shashamane, Ethiopia International Water Management Institute (IWMI), P.O. Box 5689, Addis Ababa, Ethiopia Center for International Forestry Research, Forests and Livelihoods Program, Ethiopia Office, Box 5689, Addis Ababa, Ethiopia

a r t i c l e

i n f o

Article history: Received 19 November 2011 Received in revised form 17 November 2012 Accepted 11 January 2013 Available online 12 February 2013 Keywords: Firewood Forest dependence Gender Household income Livelihoods Wealth status

a b s t r a c t Rural households across developing countries rely on diversified sources of income, and forest resources play important role in this regard. This study was designed with the objectives of assessing the contribution of forests to annual income of rural households and identifying its determinants using the case of Jelo Afromontane forest in eastern Ethiopia. It also examined the gender dimensions of forest income, and how this income varies with the wealth status of households. Key informant interview, focus group discussion and household-based questionnaire survey were used to collect data. On average, income from crop production accounted for 40.7% of the total annual household income. Forest income was second in importance, contributing 32.6%. Income from livestock, off- and non-farm activities, and woodlots accounted for 13.6%, 11.4%, and 1.7% of the total household income respectively. Firewood was the most used forest product and constituted the largest proportion (79%) of the total forest income. The contribution of forest income to the total household income varied significantly (P b 0.05) with wealth category. Forest income was more important for poor households (47.3%) than for medium (30.5%) or rich (20.2%) households. It was also more important for female headed households (58.2%) than for male headed households (29%). The gender dimension of forest income was also apparent within the household. Female members generated about four times more forest income (77% of the household forest income) than male members (23%). The sex of the household head (P b 0.01) and distance to the forest (P b 0.05) were the two determinant variables that significantly affected forest income out of the eight explanatory variables considered in the regression model. Policy to promote new forest management arrangement such as participatory forest management (PFM) in Jelo forest needs to take into account the major forest users and the types of products they depend on, and be accompanied with other poverty reduction measures so that improved forest conservation outcome will not have negative consequences on local livelihoods, particularly on poor and women, who depend most on the forest. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Rural households rely on diversified sources of income, both agricultural and non-agricultural for their livelihoods. Empirical investigations on the importance of non-agricultural sources of rural household income underline their relative importance (e.g. Cavendish, 2000; Vedeld et al., 2004), but their contributions are regularly overlooked in poverty surveys (Cavendish, 2000; McConnell, 2008). Given the increasing risk climate change and climate variability pose on agriculture, enhancing and sustaining income from non-agricultural activities would strengthen adaptive capacity of farming communities. Among the non-agricultural resources that support livelihoods of millions of people throughout the world, more importantly those in the developing countries are forests. Globally some 350 million

⁎ Corresponding author. Tel.: +251 461109900; fax: +251 461109983. E-mail addresses: [email protected], [email protected] (M. Lemenih). 1389-9341/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.forpol.2013.01.005

people are directly dependent on forests for their survival (World Bank, 2006), and many more depend on these resources to earn additional income. Forests offer the provisioning services to the poor especially during times of need; hence add to rural peoples' livelihood security (Vedeld et al., 2004; Shackleton et al., 2007). Forest products collection and sale could also support the efforts of poor households to accumulate assets and also to invest towards a more secure livelihood, for example, through educating children, purchasing agricultural inputs, or investing capital on activities that would generate additional income. On the other hand, some scholars (e.g. Angelsen and Wunder, 2003; Belcher et al., 2005) have raised doubts on the potential of forests in poverty alleviations, arguing that this potential is apparently small. This conclusion has been, however, challenged by many authors. For instance, Shackleton et al. (2008) reported employment in informal and formal forestry sector helps in moving households out of poverty. Likewise, LóPez-Feldman et al. (2007) and Tesfaye et al. (2010) demonstrated that forest income could raise the income

2

A. Asfaw et al. / Forest Policy and Economics 28 (2013) 1–7

levels of poor households to be closer to the level of the wider community. Benefits from forests, however, vary with households depending on their socio-economic characteristics (e.g. wealth status; family size in general and composition of female and male members in a household; education level, sex and age of the household head), access to forests, access to markets, institutional arrangements governing access to forests and marketing channels, and off-farm employment opportunities (Carter and Gronow, 2005; Abebaw et al., 2012). Forest income is particularly important to the poor. Generally poorer households derive greater proportion of their income from high volume low value forest products than do wealthy households. With increasing wealth, households tend to purchase forest products than collecting by themselves (Escobal and Aldana, 2003; Shackleton and Shackleton, 2006). Gender is also an important factor to consider when studying forest dependence (Ruiz-Perez et al., 2002). Gender roles can be seen at two different levels: between female and male members within a household and between male headed and female headed households. Within households, men and women often do have differing roles and responsibilities with respect to the collection, processing and marketing of forest products. In most developing countries meeting household food and fuel needs, including generating the income needed to provide these necessities, has generally been seen as the responsibility of women (Neumann and Hirsch, 2000; Prema, 2002). Men, on the other hand, are the primary harvesters of high value products such as timber that are procured deep in the forest or require hard physical labor (Clarke et al., 1996). Moreover, women in general have less access to credit (e.g. to buy farm inputs such as fertilizers and improved seeds), extension services and improved technologies (World Bank, 2008). This forces them to depend heavily on natural resources notably forests (Djoundi and Brockhaus, 2011). In sub-Saharan Africa in particular women from the poorest households obtain major source of their subsistence from a diverse set of forest products (Timko et al., 2010). A study of 25 markets in the humid forest zone of Cameroon showed that 89% of NTFPs traders were women (Awono et al., 2010). According to a study in Uttar Pradesh in India, women derive 45% of their income from forests and common lands as opposed to only 13% for men (FAO, 1996). In spite of the large body of literature on people–forest relationships, a review by Byron and Arnold (2005) raised a number of key questions that are yet to be answered. Some of these questions revolve around the characteristics of the users, the amount of benefit flows, and whether the benefit is from forest proper or trees managed outside forests. Thus, additional empirical researches on the level of people's dependence on forests, the characteristics of dependent households and factors determining their levels of dependence are still essential areas of research to be looked into across different geographic locations. The objective of this study was to determine the contribution of forest to the annual income of households around Jelo Afromontane forest in the eastern highlands of Ethiopia. The study aimed to address the following questions: • What are the major sources of income for the households in the study area? • What is the actual and relative contribution of forest income to the annual household income? • How does forest income vary with socio-economic status? And what types of forest products are particularly important to the poor? • What factors affect households' dependence on forest income? And how does gender factor (composition of females and males within a household and gender of the household head) affect dependence on and level of forest income? In line with the objective and research questions, the assumptions of the study were: (i) forest products make important contributions to the annual income of communities living around the forest, and

(ii) socio-economic parameters, including gender influence the relative importance of forest income. 2. Methodology 2.1. The study area The study was conducted in selected villages surrounding the Jelo forest in eastern Ethiopia. The forest is located between 40° 45′and 40° 52′ East longitude and 8° 55′and 9° 02′ North latitude. It is close to Chiro town, the capital of West Hararghe Zone administration (Fig. 1) some 325 km east of Addis Ababa. The altitude ranges between 1500 and 3074 m a.s.l. The area is characterized by a bimodal rainfall distribution with the average annual rainfall of 577 mm. The mean daily temperature ranges between 15.3 and 26.9 °C (National Meteorological Agency, NMA, 2009). Jelo forest represents the few remnant Afromontane natural forests in the Eastern highlands of Ethiopia. It is one of the 39 high conservation value forests identified as high priority regional forest areas in the Oromia National Regional State. The forest is also unique as a transitional forest type between the Afromontane rainforest and the dry evergreen forest. In response to severe deforestation and degradation of the Jelo forest, the Oromia Forest and Wildlife Enterprise (OFWE) in collaboration with the German agency GIZ (Deutsche Gesellschaft für Internationale Zusammenarbeit) are attempting to introduce a Participatory Forest Management (PFM) scheme that will engage communities in forest management and in return recognizes community's use rights of forests for better conservation and livelihood outcomes (Anonymous, 1990). By interviewing community members residing in the villages around the forest, attempt was made to understand the existing community–forest interaction in terms of the role the forest is playing in local livelihood strategies. The findings of the study are expected to provide essential information and additional insights that would be used in facilitating successful implementation of the PFM scheme in the study area and elsewhere. 2.2. Data collection and analysis Key informant interviews, focus group discussions and householdbased questionnaire survey were used to collect the required data. Discussions with the experts of the Hararghe Branch Office of Oromia Forests and Wildlife Enterprise (OFWE) and Chiro District Office of Agriculture revealed that the total number of kebeles (the lowest administrative unit in Ethiopia) surrounding Jelo forest was five. Out of the five kebeles, three were randomly selected for the study. From villages within each kebele, two were selected randomly. Furthermore, with the help of Government extension workers (known as development agents) and kebele administration officials, two key informants per village were identified and a total of 12 key informants were interviewed. Eight focus group discussions (with 2 women and 6 men groups) were also conducted to gather mainly qualitative information on major rural livelihood activities, forest dependence, household's risk minimization strategies and existing forest management system. The key informant interviews and focus group discussions were led by the researchers. During the focus group discussions participants were encouraged to talk freely and spontaneously but the discussions were guided and covered specific topics such as income sources, how women and men within households are engaged in various income generating activities, main types of forest products collected and the time of the year when people depend most on the forest products. The information generated was used both to guide the focus of the formal survey and also to cross check the results from the formal survey. Quantitative data were collected using the household based questionnaire. Sample households were selected using a multistage

A. Asfaw et al. / Forest Policy and Economics 28 (2013) 1–7

3

Fig. 1. Location of the study area in Ethiopia.

random sampling technique (kebeles, villages and wealth categories). The lists of all household heads residing in the selected villages were obtained from the respective kebele offices. The key informants were used in the classification of each of the households into one of the wealth categories, namely poor, medium or rich. To do so, key informants used certain criteria (Table 1). Finally, a total of 90 households, 30 households each from the three wealth categories were randomly selected to which the questionnaire based survey was administered. The questionnaire was prepared first in English, and then translated to the local language in the study area, Afan Oromo and pre-tested. It was administered using the local language, Afan Oromo, by trained enumerators under the supervision of the researchers. The questionnaire focused on household demographic and socio-economic characteristics, household assets and income sources, engagement of different household members in the different income generating activities, etc. The income assessment was based on recall for one calendar year (2008/2009), in which Ethiopian calendar corresponds to September 1st 2008 to August 31st 2009. The data collection was conducted between October 2009 and February 2010.

Table 1 Criteria used by key informants to characterize households and categorize them into wealth categories. Wealth status

Criteria

Rich

Own more than 0.75 ha of farm land, of which at least 0.25 ha is planted to khat (Catha edulis) and having access to irrigation. Own 10 or more livestocks, including milking cows, a pair of plowing oxen, equines, sheep and goats. They are able to purchase agricultural inputs (fertilizer and improved seeds) in time. They have well-constructed house. Own 0.5 to 0.75 ha of farmland (may or may not have khat field), and own between 3 and 10 livestocks, mainly cattle. They have a single but in some cases a pair of oxen. May also have a donkey, and few sheep and goats. They may have well-constructed house. Own less than 0.5 ha of land and up to 3 livestocks. They could be landless. They are unable to purchase agricultural inputs (fertilizer and improved seed), and to get oxen to timely plow their fields. They are commonly late planters, since they spend their time on other off/non-farm activities (such as daily labor, other safety-net work) to secure additional income.

Medium

Poor

2.3. Valuation method used In estimating environmental income, the study adopted household economic approach used in a number of similar studies (e.g. Cavendish, 2002). All farm products produced by a household (from crop, livestock, woodlots) were identified, their volumes estimated and multiplied by the local market price per unit volume to arrive at total farm income. Total farm income was added to forest income and to the income a household obtained from non- and offfarm activities to provide the estimate of the annual total household income during the year 2008/09. Major production costs considered relate to inputs and labor (excluding family labor). These costs were calculated and deducted from the total household income to estimate net household income. Accordingly, net income for example from crop was calculated as the difference between the total income from all crop production activities (crops and crop-residues harvested by the household) and the input costs incurred for crop production (the sum of wages paid to hired labor, costs of inorganic fertilizers, manure, improved seeds, and pesticides, rent paid on land leased in, and interest paid for farm capital borrowed). Net income from livestock farming was also calculated likewise for the six livestock species, namely cattle, sheep, goats, donkeys and chicken. Livestock income is calculated as sum of sales of livestock and their products such as milk and eggs, while costs include hired labor, feed, and veterinary costs. Forest income was calculated by estimating the total volume of forest products collected by a household and multiplied by the market price per unit volume. Forest products considered were firewood, grass, wood for construction and farm implements, tree seeds, honey and medicinal plants. Timber was not considered since cutting trees for timber is legally forbidden by the government, and none of the respondents reported income from it. No input costs were deducted from forest income as only household labor was used. Incomes from off- and non-farm activities include income from occasional labor employment, wage employment, petty trade, remittances, rent from leased out land, and different forms of aid received from NGOs or the government where people in need would receive food or cash in return for their labor services in public works such as rural road construction or maintenance, soil and water conservation activities, and forest rehabilitation. Those unable to work (e.g. elderly, and disabled) get free assistance.

4

A. Asfaw et al. / Forest Policy and Economics 28 (2013) 1–7

3. Results 3.1. Household characteristics Of the 90 study households, 20 (22.2%) were female headed and 70 (87.8%) were male headed. Only 8.9% of the respondents attended secondary school, 21.1% attended primary school, and 70% were illiterate. The age of the respondent household heads varied between 18 and 75 years, with the mean age of 38.5 years. Family size ranged between 2 and 16, with a mean of 6 persons. In terms of age composition, 52.1%, 45.7% and 2.2% of the family members were in the age ranges of 0 to 14 years, 15 to 64 years and 65 years and above respectively, indicating that substantial proportion of the population is young. The total dependency ratio was about 1.2. The landholding size of the households ranged from 0 to 3.5 ha, with the mean of 0.72 ha. Female-headed and male-headed households varied in their mean landholding sizes, which was 0.4 ha and 0.76 ha, respectively. Livestock holding ranges from 0 to 7.9 TLU, 1 and the average was 2.9 TLU. For households who cannot produce enough, food self-insufficiency is most pronounced during the months of June, July and August. Households employ different strategies to cope with food shortages. One such strategy is the collection and sale of forest products, notably firewood. 3.2. Major sources of household income and their relative contributions People in the study area are engaged in a variety of activities that comprise crop and livestock production, forest products collection, and off- and non-farm income generating activities. Income from crop production accounts for 40.7% of the total annual income of the respondent households. Forest income was the second most important and contributes to 32.6%. Income from livestock, off- and non-farm activities, and woodlots accounted for 13.6%, 11.4% and 1.7% of the total annual household income, respectively (Fig. 2). Given the vulnerability of people in the area, different aid agencies (e.g. CARE Ethiopia) and the Government support people either in kind (giving food grain and oil) mainly for the elderly and children that are not able to work or in the form of cash for those who are vulnerable but capable to work. Aid is the most important contributor (65.4%) to the total off- and non-farm income, followed by petty trade (21.9%). Income from wage employment, occasional daily labor work and rent from land leased out contributed to 4.4%, 4.0% and 4.3% of the total off- and non-farm income, respectively. 3.3. Income from various forest products and determinants of forest income levels

Fig. 2. Major sources of annual household income and their contributions in Birr in the surroundings of Jelo forest, eastern Ethiopia.

(Fig. 3). Firewood income constitutes the largest proportion (79%) of the forest income. Grass for livestock feed is the second most important product collected from the forest using cut and carry system. Focus group discussions and key informant interviews revealed that households without livestock also collect grass from the forest and sale it in nearby markets. The average contribution of grass to the forest income was 9%. Income from wood for construction and farm implements amounts to 6.9%. The collection and sale of seeds of some important indigenous tree species (e.g. Olea africana, Podocarpus falcatus, Juniperus procera and Hygenia abbyssinica) accounted for about 3.8% of the forest income. Income from honey and medicinal plants cover the remaining 1.1% and 0.2% of the forest income, respectively. Among the eight explanatory variables expected to influence level of forest income, results of the multiple linear regression revealed only two variables to significantly affect forest income. Sex of the household head (being female) positively and significantly affected the level of forest income (P b 0.01) whereas distance from the forest had negative and significant (P b 0.05) effect on the level of forest income. The age and the educational level of the household head, family size and wealth status of the household (poor being the reference category) did not significantly affect forest income levels. Though not statistically significant, forest income declined with positive changes in the education level and wealth status of the household head (Table 2). 3.4. Importance of forest income in relation to wealth category The contribution of forest income to total household income varied significantly (P b 0.05) with the wealth category. The income derived from the forest was on average Birr 3259.2, 3021 and 2442.5 respectively for the poor, medium and rich households. In

Almost all of the sample households were engaged in forest products collection. Forest income of sample households ranged from Birr 2 21 to Birr 8016 with a mean of Birr 2917.6. Firewood, grass, tree seeds, wood for construction and farm implements were the major forest products collected by the community. Medicinal plants and honey are also collected from the forest though to a limited extent. No household reported forest grazing, cutting forage trees or extracting timber. This is not unexpected as forest grazing and cutting trees are not allowed by local authorities. Focus group discussions also confirmed that forest grazing and cutting trees for timber were prohibited activities. Firewood is the most important forest product used by 88.9% of respondent households. Income of a given household from firewood could reach as high as Birr 5096, with a mean value of Birr 2304.9 1 TLU is Tropical Livestock Unit which is used to aggregate total livestock in a household. 2 Birr is the local currency in Ethiopia, and during the study one USD was exchanged for Birr 16.50.

Fig. 3. The contribution of different forest products to total forest income in Birr per year per household in the surroundings of Jelo forest, eastern Ethiopia.

A. Asfaw et al. / Forest Policy and Economics 28 (2013) 1–7 Table 2 Regression result of household total annual forest income on some selected explanatory variables (N=90). Factor

Coefficient(B)

Std. error

t-value

P-value

(Constant) Age Sex Illiterate Primary education Family size Rich households Medium households Distance

1680.240 10.873 1363.994 −132.218 −839.555 52.093 −735.315 −114.610 −234.956

976.183 13.622 502.361 520.425 581.227 68.883 416.966 378.344 110.079

1.721 .798 2.715 −.254 −1.444 .756 −1.763 −.303 −2.134

.089 .427 .008** .800 .152 .452 .082 .763 .036*

R-square = 0.276; adjusted R-square = 0.204; F = 3.85, P = 0.001; * Significant at 5% level; ** significant at 1% level.

relative terms, forest income contributes to 47.3%, 30.5% and 20.2% of the total annual household income of the poor, medium and rich households, respectively. To further identify which forest products are mostly extracted by the poor households and which ones by the rich, the relative ratios of income of rich households to the income of poor households were calculated for each of the forest products. The results show that the ratios for firewood (0.77), construction material (0.34) and medicinal plants (0.07) were less than a unit whereas the ratios for grass (1.63), forest seed (1.7) and farm implements (6.34) were greater than a unit. This illustrates that tree seeds, grass and farm implements are collected more by rich households whereas poor households depend on the forest mainly for firewood and construction materials.

3.5. Gender dimension of forest income The mean forest income generated by female headed and male headed households was Birr 8053 and Birr 10,840, respectively. Though forest income in absolute value was high for male headed households, its relative importance to total household income was much higher for female headed households (58.2%) than male headed households (29%) (Table 3). The gender dimension of forest income was also apparent within the household. The forest income generated by female members (Birr 2341 or 77% of the total household forest income) varied significantly from the amount generated by male members. The income generated by female members was almost four times greater than the amount generated by male members, which was Birr 566 or 23% of the total household forest income. Female members were involved mainly in firewood collection and to a certain extent in cutting and carrying grass for livestock feed. About 81.5% of firewood collection task was accomplished by female members of the household only, whereas males only and both male and female members undertook respectively, 2.5% and 16% of the firewood collection tasks conducted by a household. About 96.2% and 3.8% of the income from firewood and 5.3% and 94.7% of income from grass were generated by female and male household members, respectively. The contribution of male members were more pronounced for incomes from wood for

5

construction and farm implements, forest seeds, medicinal plants and honey (Table 4). 4. Discussion This study shows that products from Jelo forest make significant contribution to the income of households living adjacent to it. But such contributions have not been properly documented and accounted for in poverty analysis and rural development planning in Ethiopia (Tesfaye et al., 2010). Forest income is the second most important source of household income, contributing to about onethird (32.6%) of the total annual household income even without inclusion of income from timber. Informal discussion reveals that Jelo forest is also exploited for timber. As this is considered illegal, and hence punishable, it was not reported by any of the respondents. If this income from timber was accounted, total forest income would have been higher, illustrating further the importance of forests for household income in the study area. Lack of information on income from timber could be considered as limitation of this study. Despite this limitation, the findings of this study are comparable with studies conducted in other forested areas of Ethiopia (e.g. in Dindi district (Mamo et al., 2007), in the Bale highlands (Tesfaye et al., 2010), in the Liban Zone of Somali Region (Lemenih et al., 2003) where forest income contribution was respectively 39%, 34% and 32%. In Shendi district in Zimbabwe, Cavendish (2000) reported that forest income accounted for about 35% of total annual household income, which is also comparable to the present study. Firewood, grass for livestock, wood for farm implements and forest seeds were the four major forest income sources in their descending order of importance. However, firewood dominates accounting for 78.9% of total forest income. This level of contribution from firewood is much higher than the 59% reported by Mamo et al. (2007) for Dindi district and 55% for the Bale highlands (Tesfaye et al., 2010). A global assessment also shows the dominant role of fuelwood in forest environmental incomes for the rural poor (Vedeld et al., 2004). Mean forest income per household varied with wealth status. Poor households earned significantly more income. This may be due to the fact that poor households have fewer assets in terms of land, livestock, and cash to generate more income from agriculture. Thus they tend to depend more on forests. This high dependency of poor households on forests is in agreement with findings of other studies (e.g. Godoy and Bawa, 1993; Shackleton and Shackleton, 2006). Poor households depend on firewood and construction materials to a large extent, and on medicinal plants and grass to a lesser extent. Most of these products are sold in the nearby markets. Rich households, on the other hand, depend mainly on grass for livestock feed, wood for farm implements, tree seeds and honey. In terms of determinants of total forest income, the sex of the household head (being female) was positively and significantly related to household's total forest income. This implies that a female-headed household is more likely to be forest product collector than a male-headed household. This result is in line with study carried out by Mohamed (2007) in Bench Maji and Sheka zones of southern Ethiopia and by Babulo et al. (2009) in the highlands of Tigray, northern Ethiopia. Forest income shows significant reduction with increasing distance to the forest. This negative relationship is

Table 3 Major income sources and their respective contributions in value (Birr) and percentage to total household income disaggregated by the gender of household heads in the surroundings of Jelo forest, eastern Ethiopia. Household head

Female Male Paired t-test of income level (P value)

Income sources and their contributions Total income

Crop income

Forest income

Livestock income

Off/non-farm income

Income from woodlots

8053 (100%) 10,840 (100%) 0.000

1542 (17%) 4624 (44%) 0.000

4526 (58%) 2687 (29%) 0.000

778 (7.5%) 2107 (15%) 0.000

1205 (17.5%) 1351 (10%) 0.000

0 167 (2%)

6

A. Asfaw et al. / Forest Policy and Economics 28 (2013) 1–7

Table 4 Forest products collected and their respective income levels by gender of family members in the surroundings of Jelo forest, eastern Ethiopia. Forest product

Firewood Grass Wood for construction and farm implements Tree seeds Honey Medicinal plants

Absolute income (Birr)

Proportion (%)

Female

Male

Female

Male

Income level

Proportion

2155 332 0 0 0 0

112 18 56.7 55.7 35.4 2.8

96.2 5.3 0 0 0 0

3.8 94.7 100 100 100 100

19.2 18.4 0 0 0 0

25.3 0.06 0 0 0 0

also in agreement with the report of Mamo et al. (2007) for Dindi district in Central Ethiopia and Abebaw et al. (2012) for Metema district in northwestern Ethiopia. Regarding the gender dimension of forest income, marked differences between males and females were observed both across and within households. Forest income is more important for female-headed households than male-headed households. This has to do with limited asset level of female-headed households (e.g. their land size is half that of male-headed households). This result is in line with the findings of a number of other studies (e.g. Cavendish, 2000; Shackleton and Shackleton, 2006) that also concluded that forests contribute more to female-headed households than to male-headed households. Also, in a given household, it is primarily the female members who are the major gatherers of forest products. This finding is comparable with that of Tessema (1997) who reported that 83% firewood collectors in Chiro district (eastern Ethiopia) were females. Similarly the mean firewood income contribution by female and male household members was 96.2% and 3.8% respectively. This also indicates that meeting the household energy demand in the study area is still the responsibility of female household members. But most income (95%) from grass was mainly obtained by male members. Female household members are rarely involved in cutting and transporting grass, generating only 5% of income from it. The overall mean household forest income contribution by female members remained much higher than that of male members. The greater dependence of female-headed households and female members on forest income more than male headed and male members of a household emanates mostly from the fact that they lack the necessary information, skill, capital, land and other assets to be engaged in intensive agricultural production or in non-farm activities and to increase their income. 5. Conclusions and implications The findings of the study illustrate that for communities residing around Afromontane forest in Eastern Ethiopia forest income constituted a third of the total annual household income and was the second most important income source, next to crop. Gender and wealth status influenced the types of forest products used by the households as well as the income levels generated from these products. Male-headed households earned more in terms of absolute value of forest income but relative importance of forest income to total household income was much higher for female headed households. Even in male-headed households, female members generated significantly more forest income than male members showing the gender dimension of forest income and its particular importance to females in the community. The study results are important in informing forest policy and management practices, including the new PFM scheme in Ethiopia, as the country has initiated a national project to scale up PFM on millions of hectares of state forests. The experience of PFM was made possible thanks to the policy provisions that encourage participation of communities in management and use of forests. Article 4 (3) (b) and Article 4 (3) (e) of the national forest policy issued in 2007 and Article 10 (3) and Article 11 (6) of the Forest Proclamation

Ratio female/male

enacted in 2007, allow for designing and implementing participatory management strategy that can be utilized by the local communities in managing protected forests and as per the management plan, to be drawn by appropriate government bodies, the right to use nontimber forest products such as honey, species, wild coffee, fodder, and dried and fallen woods. But the national forest act and the detailed guideline that follows the Act are yet to be finalized but key documents to facilitate smooth implementation of the provisions of the forest policy and proclamations on the ground. Therefore, the national forest act and the detailed forest guidelines to be issued need to recognize that communities are not uniform in their dependence on and use of forests. The particular dependence of the poor and women on forests needs to be emphasized and provisions need to be made in clearer terms to make sure that these segments of the community are included in the process of PFM planning and implementation in general and in defining the forest management plan in particular. Current practices focus on involving communities as an entity and do not specifically make efforts to empower and ensure the participation of the poor and women in defining the process and outcomes of PFM. Unless this recognition is made and especially tailored support is provided to this segment of the community, elites within the community are likely to steer the process and take the lion's share of the benefits to be accrued through PFM. Limiting the use of forest products to NTFPs, as stipulated in the forest policy, is also the other policy related challenge that continues to undermine economic incentives for communities to be engaged in responsible management of forests. Besides policy-related constraints, PFM practices in Ethiopia are reportedly showing tendency towards forest protection than livelihood enhancement (Temesgen and Lemenih, 2012). The participation of women and poor households in PFM processes remains marginal. Our findings show that forest products are the second most important sources of household income, and the relative contribution of forest income was much higher for the poor and women. Thus focusing mainly on forest protection will undermine forest income of particularly the poor and women, and may in the long run further marginalize them. Thus the national forest act and the national forest guideline yet to be issued should clearly recognize the dependence of community members on forest resources, the particular dependence of the poor and women, and should include provisions to ensure inclusive and genuine participation of the poor and women in decision making in forest management so that PFM contributes to achieving both conservation and livelihood outcomes.

Acknowledgments The authors thank the staff of the Haraghe Branch of Oromia Forest and Wildlife Enterprise, and the GIZ advisers to the PFM project in Jelo Forest for their collaborations. We also thank Chiro District Agricultural and Rural Development Office and the District Land and Environmental Protection Office for providing us with the necessary secondary data. We are grateful to the communities and developmental agents in the area for their collaboration during the study. We also

A. Asfaw et al. / Forest Policy and Economics 28 (2013) 1–7

thank Wondo Genet College Forestry and Natural Resources and CIFOR for providing financial and technical support for the study. References Abebaw, D., Kassa, H., Kassie, G.T., Lemenih, M., Campbell, B., Teka, W., 2012. Dry forest based livelihoods in resettlement areas of Northwestern Ethiopia. Forest Policy and Economics. http://dx.doi.org/10.1016/j.forpol.2012.02.002. Angelsen, A., Wunder, S., 2003. Exploring the forest–poverty link: key concepts, issues and research implications. CIFOR Occasional Paper No. 40.Center for International Forestry Research, Bogor, Indonesia. Anonymous, 1990. Management Plan of Jelo Forest (1990). Tehasas1982–Gumbot 1986 Ethiopian Calendar (EC.).Assebe Teferi, Western Hararge. Awono, A., Ndoye, O., Preece, L., 2010. Empowering women's capacity for improved livelihoods in non-timber forest product trade in Cameroon. International Journal of Social Forestry (IJSF) 3, 151–163. Babulo, B., Bart, M., Fredu, N., Tollens, E., Nyssen, J., Deckers, J., Mathijs, E., 2009. The economic contribution of forest resource use to rural livelihoods in Tigray, Northern Ethiopia. Forest Policy and Economic 11, 109–117. Belcher, B., Ruíz-Pérez, M., Achdiawan, R., 2005. Global patterns and trends in the use and management of commercial NTFPs: implications for livelihoods and conservation. World Development 33, 1435–1452. Byron, N., Arnold, J.E.M., 2005. What future for the peoples of the tropical forests. In: Sayer, J. (Ed.), Forestry and Development. Earthscan, London, pp. 145–168. Carter, J., Gronow, J., 2005. Recent experience in collaborative forest management: a review paper. CIFOR Occasional Paper No. 43.CIFOR, Bogor, Indonesia. Cavendish, W., 2000. Empirical regularities in the poverty–environment relationship of rural households: evidence from Zimbabwe. World Development 28, 1979–2003. Cavendish, W., 2002. Qualitative Methods for Estimating the Economic Value of Resource Use to Rural Households: Uncovering the hidden harvest. In: Campbell, B.M., Luckert, M.K. (Eds.), pp. 17–63 (London). Clarke, J., Cavendish, W., Coote, C., 1996. Rural households management and miombo woodlands: use value and management. In: Campbell, B.M. (Ed.), The Miombo in Transition: Woodlands and Welfare in Africa. CIFOR, Bogor, pp. 101–136. Djoundi, H., Brockhaus, M., 2011. Is adaptation to climate change gender neutral? Lessons from communities dependent on livestock and forests in northern Mali. International Forestry Review 13, 123–135. Escobal, J., Aldana, U., 2003. Are non-timber forest products the antidote to rainforest degradation? Brazil nut extraction in Madre de Dios, Peru. World Development 31 (11), 1873–1887. FAO, 1996. Forest dependent survival strategies of tribal women: implications for joint forest management in Andhra Pradesh, India. In: Yadama, GN., Pragada, BR., Pragada, RR. (Eds.), FAO Regional Office for Asia and the Pacific, Bangkok, Thailand. Godoy, R.A., Bawa, K.S., 1993. The economic value and sustainable harvest of plants and animals from the tropical forest: assumption, hypotheses and method. Economic Botany 47, 215–219. Lemenih, M., Abebe, T., Olsson, M., 2003. Gum and resin resources from some Acacia, Boswellia and Commiphora species and their economic contribution in Liben, Southeastern Ethiopia. Arid Environment 56, 146–166.

7

LóPez-Feldman, A., Mora, J., Taylor, J.E., 2007. Does natural resource extraction mitigate poverty and inequality? Evidence from rural Mexico and a Lacandona Rainforest community. Environment and Development Economics 12, 251–269. Mamo, G., Sjaastad, E., Vedeld, P., 2007. Economic dependence of forest resources: a case from Dendi District, Ethiopia. Forest Policy and Economics 9, 916–927. McConnell, R., 2008. Links between national forest programmes and poverty reduction strategies. Forestry Policy and Institutions Working Paper -Food and Agriculture Organization, Rome (Page vii + 45 pp.). Mohamed, A. 2007. The contribution of non-timber forest products to rural livelihood in southwest Ethiopia. MSc. Thesis, Wageningen University and Research Center National Meteorological Agency (NMA), 2009. National Meteorological Agency, Adama branch, Chiro site. Neumann, R.P., Hirsch, E., 2000. Commercialization of non-timber forest products: review and analysis of research. CIFOR, Bogor, Indonesia. Prema, G., 2002. Women's role and contribution to forest-based livelihoods. Paper reported to UN. New Delhi and India. Ruiz-Perez, M., Ndoye, O., Eyebe, A., Ngono, D.L., 2002. A gender analysis of forest product markets in Cameroon. Africa Today 49, 96–126. Shackleton, C.M., Shackleton, S.E., 2006. Household wealth status and natural resource use in the Kat River Valley, South Africa. Ecological Economics 57, 306–317. Shackleton, C.M., Shackleton, S.E., Buiten, E., Bird, N., 2007. The importance of dry woodlands and forests in rural livelihoods and poverty alleviation in South Africa. Forest Policy and Economics 9, 558–577. Shackleton, S.E., Campbell, B., Lotz-Sisitka, H., Shackleton, C.M., 2008. Links between the local trade in natural products, livelihoods and poverty alleviation in a semiarid region of South Africa. World Development 36, s05–s526. Temesgen, Z., Lemenih, M., 2012. Gaps Assessment and Analysis of Participatory Forest Management Activities in Ethiopia. MoA, Addis Ababa, Ethiopia. Tesfaye, Y., Roos, A., Campbell, B., Bohlin, F., 2010. Forest incomes and poverty alleviation under participatory forest management in the Bale Highlands, Southern Ethiopia. International Forestry Review 12, 66–77. Tessema T. 1997. Dependency on Forest for fuel wood and the prospect of tree growing by peasant community: the case study at Jelo forest area, western Hararge, East Ethiopia. Msc Thesis. Hawassa University, Wondo Genet College of forestry and Natural resource; Wondo Genet, Ethiopia. Timko, J.A., Waeber, P.O., Kozak, R.A., 2010. The socio-economic contribution of nontimber forest products to rural livelihoods in Sub-Saharan Africa: knowledge gaps and new directions. International Forestry Review 12 (3), 284–294. Vedeld, P., Angelsen, A., Sjaastad, E., Berg, G.K., 2004. Counting on the environment: forest incomes and the rural poor. Environmental Economics Series Paper No. 98. The World Bank, Washington DC. World Bank, 2006. At loggerheads? Agricultural expansion and poverty reduction in tropical forests. World Bank Policy Research Report. The World Bank (http://go. worldbank.org/TKGHE4IA30). World Bank, 2008. Gender in Agriculture Sourcebook. The World Bank, Washington, D.C. (http://go.worldbank.org/YQ4LDN9AB0).