Effects of Large-Scale Agricultural Land Investment on Local Communities Livelihoods: Evidence from Bambasi Woreda, Western Ethiopia

Shafe Zelalem Gasisa; Mihret Fentahun Yeneneh; Teha Romanu Benti

doi:doi:10.11648/j.ajere.20240902.11

Research Article |

| Peer-Reviewed

Effects of Large-Scale Agricultural Land Investment on Local Communities Livelihoods: Evidence from Bambasi Woreda, Western Ethiopia

Shafe Zelalem Gasisa, Mihret Fentahun Yeneneh

, Teha Romanu Benti^*

Published in American Journal of Environmental and Resource Economics (Volume 9, Issue 2)

Received: 9 April 2024 Accepted: 24 April 2024 Published: 10 May 2024

Views: Downloads:

Download PDF

Share This Article

Twitter
Linked In
Facebook

Abstract

Most of the African countries including Ethiopia are often known by problem of large scale agricultural land investment effects on the local community where land is being transferred to investment. Even though several efforts made so far to solve the overall effects of large scale agricultural land investment situation, the challenge is still widespread problem in Ethiopia. Hence the study project's purpose was to find out how large agricultural investments in Bambasi Woreda, Western Ethiopia impact the livelihoods of the surrounding populations and examine the local communities' participation in the large scale agricultural investment. In order to attain these objectives, data were collected from 330 randomly selected households in four purposively selected kebeles of the district for both control groups and treatment groups. The sample size was chosen using a multistage stratified random sampling technique. Both qualitative and quantitative data gathering techniques and instruments were employed in the study. Besides, the instruments utilized to collect the data were observations, focus groups, interviews, household surveys, and document reviews. Data was analyzed using both descriptive statistics and econometric methods. The study shows that out of the total sampled respondents 30 of the treatment group and 13 percent of control group reveal that the project provided opportunity in terms of employment opportunity, technology transfer, utilization of agricultural inputs, changing the working culture of the community and productivity. The chie square value shows there is statistical significance among treated and control group on opportunity investment provided for the household and community. A binary logit regression model was used to describe how large-scale agriculture land investment affected the local community's standard of living. The findings indicated that only six variables were found to be significant out of the characteristics that were expected to influence local community employment in large-scale agricultural land investment projects. These includes household's educational accomplishment, size of HH, occupation of HH, Loss of useful land due to investment Project and technology transfer significantly and positively affected the employment opportunity in large scale agricultural investment projects, whereas the distance of a household's home from an investment project has a negative impact. Large-scale agricultural investments have a detrimental influence on household wealth accumulation and income, according to the estimation results of the average treatment effects on the treated. The management and implementation of land transfer for large-scale agricultural investment projects is inadequate, lack of openness, absence of community consultation, natural forest degradation, socio-economic and ecological effects must be carefully considered before transferring the land for large-scale agricultural investment.

Published in	American Journal of Environmental and Resource Economics (Volume 9, Issue 2)
DOI	10.11648/j.ajere.20240902.11
Page(s)	20-39
Creative Commons	This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.
Copyright	Copyright © The Author(s), 2024. Published by Science Publishing Group

Keywords

Effects, Large Scale Agricultural Land Investment, Control Group, Treatment Group and Livelihoods

1. Introduction

Globally, large-scale land attainments, sometimes known as "land grabbing," have attracted attention from the public ever since nongovernmental organizations published media reports, articles, and research studies about the subject in 2007 and 2008. The drive for global commercial property agreements by foreign managements, private businesses, and asset moneys is brought to light by land grabs. By the end of 2020, these land investment attainments had enclosed more than 56 million hectares of agricultural land. Out of this around 4.5 million ha of land available for commercial land investment in Ethiopia

[1]

. Advocates highlight the economic benefits for surrounding communities, which will profit from the earnings produced from the land's sale or lease as well as the employment that will result from it. Promoting the essential expenditures on modern technology and infrastructure may offer additional benefits. New agricultural investments generally have the ability to contribute to the provision of the necessary conditions for sustainable development, according to

[2]

A large amount of land is being invested in as part of the Ethiopian government's development agenda. Over 75% of the work force, 40% of GDP, and 80% of exports are derived from agriculture, which is the foundation of the country's economy. Contrarily, Ethiopia experiences a significant yearly food deficit and is known for its ongoing food insecurity. Highland plots are small, fruitful, and dependent on erratic rainfall. Pastoralism, agro-pastoralism, or shifting agriculture makes up the majority of rural lowland livelihoods. Rapid agricultural change is planned, involving both smallholder sector innovation and the expansion of large-scale commercial agriculture sectors. The lowlands of Oromia, Gambela, SNNPR, Afar and Somali regions, and Benishangul Gumuz are home to the government's land bank and largest land holdings. Minority ethnic groups regard the land in these areas as vacant, notwithstanding claims that it is a part of pastoralist or shifting agricultural land use systems. The government of some places, such Benishangul Gumuz, Gambela, and Somalia, is implementing villagization plans, which entail relocating populations to create room for better service delivery.

Significant agricultural investments should not jeopardize the community, but rather strengthen it. This covers local and indigenous populations, food security, sociocultural norms, political and human rights, and the ability to access resources such as land. The African governments are provided with clear recommendations by the CAADP

[3]

on how to attain food security, evidence, technology, physical resources, and financial funds in order to facilitate community' access to the global market. The administration hasn't given any of these suggestions much thought thus far, although the reality seems to be different

[4-6]

. Developing agricultural land is meant to be the aim of this endeavor. Only 14.6 million of Ethiopia's 111.5 million hectares are currently utilized for agriculture, out of the country's total area of 74.3 million. Just 1 million hectares are being irrigated, despite the potential for 4.3 million hectares

[7]

Land that the Ministry of Agriculture and Rural Development (MOARD) was unable to coordinate and attract capital-intensive foreign direct investment to the various regions of the country made up the majority of the land that was handed to the federal land banks. The entire amount of investment land supplied to the Federal Land Bank was 3,169,352ha, divided among Amhara (420,000ha), Afar (409,678ha), Benishangul Gumuz (691,984ha), Gambella (829,199ha), Oromia (1,057,866 ha), and SNNP 180,625ha

[8]

. Under Ethiopia's federal political system, which is based on ethnic regional governance, the Benishangul Gumuz area transferred 691,984 hectares of investment land to the federal land banks. Both domestic and foreign investment increased significantly in the region between 2007 and 2019, with an estimated 316,415.9 hectares of land transferred to 531 agricultural investors.

The primary goal of the study is to determine how the locals of Bambasi Woreda, in the Ethiopian regional state of Benishangul Gumuz, are affected by a significant agricultural investment project. With a total wealth of 283,837,725.00 million Ethiopian birr invested, there are 72 large scale agricultural land investment projects in the woreda. Of those investors, 34 large-scale agricultural investments totaling more than 250ha of land have been taken as large-scale land transfers in the area (BGRSIO, 2023). However, the Report Office (2023) notes that there is extreme land grabbing in the study area, with land being transferred to investors without any consideration for how this would affect the local community's means of subsistence. Nevertheless, the specifics of the indigenous community livelihoods in the woreda or region remain unknown. Therefore, this study was made to examine how large-scale agricultural land investment initiatives affected the livelihoods of the local communities in Bambasi woreda of BGRS, Western Ethiopia.

1.1. Statement of the Problem

Land is the resource that is necessary for all of the socio- economic and ecological aspects of local communities' life. Sufficient access to land and related resources, therefore, constitute the foundation of rural communities' livelihoods and offer significant social and economic advantages. Many smallholder communities have historically relied heavily on the question of access to land and related natural resources for their subsistence. The topic is receiving more attention these days because of the continuous intense competition for this vital resource, which involves many different actors in the agricultural investment sectors. Since there are differing opinions about how "Large Scale Agricultural Investment" would affect the native population in the nations where land is being transferred, this is one of the most argued and hot topics of our today. Thus, the question was the focus of continuing discussions among researchers, lawmakers, and policy leaders, among others.

Due to the increased attention being paid to the agricultural sector globally and in particular to the acquisition of large tracts of farmland, a number of private and foreign investors from a variety of countries have been involved in the acquisition of large amounts of land in sub-Saharan Africa. Large areas of land in less developed nations, particularly in sub-Saharan Africa, have been acquired by governments and businesses in recent years with the goal of producing food crops and biofuels for export

[9]

. Similar to other countries, Ethiopia has attracted both foreign and domestic investors, who have some of the blame for improper land investments that fail to protect the social, economic, and environmental demands of the local communities. Governments create favorable conditions for investors to invest in various parts of the region, particularly specific sections of BGRS, in order to draw in both domestic and foreign capital. The primary goals of large-scale agricultural land investment in the study areas are to address the issue of food security, generate employment opportunities, reduce poverty, provide infrastructure, and transfer technologies to small-scale farmers.

Reviewing the results of other studies carried out in various locations has shown that numerous agricultural investments have caused rural household communities to be evicted and displaced, affecting their local means of subsistence, social and economic issues, and access to resources and infrastructure.

However, 1.19 million hectares of Ethiopian agricultural land were leased to both domestic and foreign investors

[10-12]

. While many people are impacted, few are aware of the conditions and benefits on those investments. Researchers looked at how different projects in various regions of Ethiopia affected the local communities as a result of LSAI. For example,

[2, 13, 14]

, concentrate on the interaction between investors and the local population, the consequences of land transactions on the socioeconomic landscape, and the environmental effects on the livelihoods of the local inhabitants.

In the specific research region, there is a dearth of attention paid to employee involvement and contributions that alter workplace cultures, job opportunities created for the community, and technology transfer to the local community that improves productivity and addresses issues with food security. As a result, the problem that spurred research on the "impact of LSAI on the livelihoods of the local community" in the study areas was the possibility of numerous investment projects, especially in the field of agriculture, being made in the woreda, and the rise in land grabbing concerns being more of a problem than a chance for the local community to improve its assets and income. Therefore, research study need to close the gaps on the contributions generated by LSAI to the socioeconomic well-being of the local communities, the availability of infrastructure, and the enhancement of asset growth and income in the studied area.

1.2. Objectives of the Study

The general objective of the study was to examine the effect of large-scale agricultural land investment on the livelihoods of the local communities. The specific objectives are as follows:

1) To examine the local communities' participation and opportunities generated by large-scale agricultural land investments

2) To examine the effects of large-scale agricultural land investments on Asset growth of local communities

2. Methodology of the Study

2.1. Description of the Study Area

The study was carried out in Western Ethiopia's Bambasi woreda. Benishangul Gumuz regional states comprise Bambasi woreda, which is bordered to the east in part by Oromiya region, to the west by Assosa woreda, to the south by a portion of Maokomo special woreda, and to the north by Odablglidu woreda. There were 62,693 people living in the woredas overall, with 13,389 men and 10,65 women making up the household population. The capital of the woreda, Bambasi Town, is the head office. It is situated at 09017'- 12006' North Latitude and 34010'-37004' East Longitude, with an elevation of 580–2730m above sea level. Assosa, a regional town, is 42 km away, while Addis Ababa is 662 km away.

There are 38 Kebeles in the woreda, and nine of those Kebeles are home to the investment projects. The four purposefully chosen Kebeles—Wombselam, Shobergushi, and Amebaa 16 and 27—where large-scale agricultural investments (LSAI) were owned and no large-scale agricultural investments projects were carried out, respectively, were the subject of the investigations. Seventy-two investment projects, all related to agriculture, are presently under implementation (Table 1). The study's main goal is to find out how those agricultural investment initiatives affect (negative or positive) local populations' livelihoods, particularly those of the indigenous population.

Table 1. The distribution of investment projects in the Bambasi Woreda.

Kebele	No. of investment projects	Land transfer to investment projects in ha.	LSAI projects
Wombselam	13	3872ha	8
Garabichwollega	12	2761ha	6
Budaselga	9	2147.4ha	3
Mustia	5	1424ha	3
Idadabus	3	625ha	0
Shobergushi	19	6176.5ha	8
Bushmakargagi	7	1654.6ha	4
Jmasta	1	250ha	1
Shbora	3	572ha	1
Total	72	19,482.5ha	34

Sources; - (BGRSIO data, BWEFLA office data, and BWARD Office post harvesting report, 2017)

2.2. Research Method

Econometrics Model Analysis Propensity score matching (PSM) analysis tools are suitable for the study based on the type, analysis, and interpretation of the surveyed data. The PSM eliminates bias resulting from observable variables by balancing covariates across the "treatment" and "control" groups, improving regression's capacity to produce precise causal estimations. Matching the treatment variable employment (E) as treatment and control group, the PSM model matches on the probability of getting the conditional probability of the treatment and being treated. The purpose of this group comparison is to assess how large-scale agricultural investment affects the creation of employment opportunities for the livelihoods of the treated groups.

2.3. Data Type and Sources

Both quantifiable and qualitative data types were used. The quantitative data’s are data that collected through organized and semi organized interview of household. The qualitative data are data collected through key informant interview, focus group discussions, and observations and filed notes.

Informants such as community household surveys, traditional community elders, government officials (bureau head, directorates and experts), owners of investment projects, project managers, extension workers (development agents) and household heads impacted by large-scale agricultural investment projects are the main sources of information for this study. In order to improve the data, secondary data was also gathered and acquired through the examination of records, published works, and reports from relevant industries.

2.4. Target Population

The local community, community households, and the community with residences close to investment projects, community those get any opportunity from investment projects and large-scale agricultural investments (investors) are the target audiences for this study.

2.5. Sampling Technique

The research was carried out in Benishangul Gumuz Regional State (BGRS) in the Bambasi woredas. The size of the household sample from purposefully chosen Kebeles was determined using the basic random sampling technique. Based on the random sampling percentage to their population members, the sample size from the treated groups that receive employment opportunities and do not have jobs in the large-scale agricultural investment was selected. In the study, development agents from the purposefully selected Kebeles and local administrations were consulted before ahousehold survey of 330 randomly selected families was conducted. Using Yamane's formula

[15]

, the samples from the treatment group were 102 households from 460 families in Wombselam Kebeles and 58 households from 261 households in Shobergushi Kebeles.

n =

\frac{N 1}{1 + N 1 (e) ²}

n = the number of sample size of the HHs selected

N1= number of the HHs from purposively selected Kebeles

e = 0.07 is the proportion of the imitations of sampling error that can be tolerated

n =

\frac{721}{1 + 721 (0.07) ²}

= 160 from the total 721 HH of the purposively selected Kebeles, HH sampled for the survey are 160.

The investment projects having large-scale agricultural investment projects (LSAIP) are also sample for the study to investigate the impacts of large sale agricultural investment on the local community’s livelihoods.

Table 2. HH Survey samples in each of the study areas in treatment groups.

Woredas	No_ of Kebeles selected	Total no_ of households	Sampled of HH
Bambasi	Wombselam	460	102
Bambasi	Shobergushi	261	58
Total sample	No_ of household	721	160

In the case of control group, 170 households randomly selected from purposively selected two Kebeles where no large-scale agricultural investments projects practiced after consultations with the local administrations and development agents using Yamane’s formula.

n =

\frac{N 1}{1 + N 1 (e) ²}

n = the number of sample size of the HHs selected

N1= number of the HHs from purposively selected Kebeles

e = 0.07 is the proportion of the imitations of sampling error that can be tolerated

n =

\frac{764}{1 + 764 (0.07) ²}

= 170 from the total 764 HH of the purposively selected Kebeles, HH sampled for the survey are 170

Table 3. HH samples for control group in each of the study areas.

Woredas	No_ of Kebeles selected	Total no_ of households	Sampled of HH
Bambasi	Amebaa 16	451	106
	Amebaa 27	313	74
Total sample	No_ of household	764	170

2.6. Methods and Instruments of Data Collection

Structured and semi-structured interviews, a household survey, interviews with key informants, focus group discussions (FGD), observations, and document reviews were all employed in the study.

Structured and semi structure interview; - Researcher used the structured and semi structure interview to extract detailed information from interviewee. In general, the main drive of conversation is to get related information from sample HH head and investment project owner/project manager.

Household Survey; the primary economic data collection instrument for the project; household characteristics, assets, income, and additional contextual factors are among the data gathered. The size, sex, age, education level, and economic activity of the household are among its qualities. The possession of land, ownership of animals, a place to live, furniture, appliances, money, and savings are all considered household assets. Crop production, livestock, off-farm self-employment (non-farm activities), off-farm wage employment, irrigation, and other activities are the sources of income collected for the household.

Focus group discussions (FGD);- is use to gather qualitative facts on fluctuations in local livings connected to LSAIinvesting near to the area of the local communities. The representatives of the selected households in each Kebeles were composed and questions highlighting on changes in right to use of land, forest and water resources, completion and conflicts over the use of local properties, fluctuations in local markets, business, infrastructures and facilities, and fluctuations in production methods and skills.

The key informants; in the context of the studies the key informant interviews are 12 large-scale agricultural investment projects, traditional community elders, Kebeles agricultural extension lead workers, woredas and regional environment, forest and land administration bureau and regional investment office experts/directorate and bureau head.

Document: - the review of official papers and policy, rules and regulation documents, regional and woredas investment facts and reports from comparative sectors. The relevant data mainly collected from Woreda administration, Bureau of environment, forest and land administration, Bureau of agriculture and regional investment office. The internet/website is also potential sources of information for the study.

Observations; - is conduct as part of the felid work to observe in and around the investment projects through taking filed notes. It comprise on the observations of land allotted for the investments and its bounders with local landholding, communal grazing land, walkways for local people and livestock, infrastructures and other services. The other observation made within the study’s areas of field works are infrastructure made by privet investors and the addressed infrastructure to the community.

2.7. Data Analysis

Econometric models and descriptive statistics are used in the analysis of the gathered data. Descriptive analysis is used to study the raw data, which facilitates understanding, interpretation, ordering, and rearranging of the data. Given that PSM is the treatment, an economic model was utilized to create the comparison groups according to the likelihood of participation. PSM model used to look into how a binary treatment affects an observation's result. Matching the treatment variable employment (E) as treatment and control group, the PSM model matches on the probability of getting the conditional probability of the treatment and being treated.

p (x) =

Pr (E=1/x) = Pr (E=0/x)

where P(x) ix the outcome, E=1 one or more HH members employee participate in LSAI, E=0 HH do not participate as employee member in LSAI, x is asset of observed characteristics. The researcher used binary logit model to analysis the data.

Y = Xβ + ε

Y- is the outcomes variable, x- is asset of observed characteristics,

β

- is parameter of interest and

ε

- is an error term that reflect the unobserved characteristics that affect the outcomes.

2.7.1. Descriptive Analysis

To assess the data, descriptive statistical techniques like variance and mean were used. The unprocessed data obtained from the completed forms was analyzed, categorized, reorganized, encoded, and entered into Mex before being imported into Stata. The relationship between land taken due to investments and food self-support, employment opportunities created, infrastructure developed, and social service delivery as a result of LSAI expansion on the study areas were also tested using statistical correlations and regressions.

2.7.2. Econometric Analysis

Propensity score matching (PSM), is an econometric model that was chosen to analyze the non-parametric techniques to balance covariates between the "treatment" and "control" groups, it eliminates bias resulting from observable variables and enhances regression's capacity to produce precise causal estimates. By matching users who are equivalent in terms of their observable qualities, both participants and non-participants are to be created using the propensity score matching approach. The models were also employed to look at how a binary treatment affected an observation's outcome. Matching the treatment variable employment (E) as treatment and control group, the PSM model matches on the probability of getting the conditional probability of the treatment and being treated. The purpose of this group comparison is to assess how large-scale agricultural investment affects the creation of employment opportunities for the livelihoods of the treated groups. Let the income for participants and non-participants, respectively, be represented by YiT and YiC.

∂i = Yi^T- Yi^C(1)

∂i = change in outcome as result of treatment or change of income for participant in the program

Yi^T= outcome of treatment (income of i^th household when one or more household members are get employment opportunity from LSAI); Yi^C= outcome of treatment (income of i^th household when one or more household members are does not participates as employment opportunity from LSAI).

The above will be questions expressed i causal effect notational form, by assigning Di= 1 as treatment variable takes the value of 1, if the household participated as treatment (get employment opportunity) and 0 otherwise. Thus the average treatment effect of household i can be written as:

ATT = E (Yi^T|D=1)-E (Yi^C|D=0)(2)

Where ATE, average treatment effect, which is the effect of treatment on income, E (yi^T/D= 1);

average outcomes for household, with treatment, if one or more of the household get employment opportunity from large scale agricultural investment (D= 1). E (Yi^C/D=0); average outcome of untreated, when the households are not participate as employee in LSAI, (D=0).

To measure the Average Effects of Treatment on the treated (ATT) for the sample can be formulated as:

ATT = E (Yi^T–Yi^C)| D=1) = E (Yi^T|D=1) – E (Yi^C|D=1)(3)

The evaluation problem in estimation of impact is that it is impossible to observe persons for with and without treatment at the same time. While the post intervention, the outcome E (Yi^T|D=1) is possible to observe.

2.8. Variables

Independent Variables

1. Household head sex- It is a dummy variable with value of 1 for male, other wise 0. It is expected that relatively male head of household will participating in LSAI as employee and get new technology from investors than female head of household.

2. Household head age- It is continuous variable measured in year. It is expected that younger families will participating in large-scale agricultural investment as employment, getting new technology and transferring to investors than older generation.

3. Household size- It is a continuous variable; the number of family size live in the same household affects household engagement as employee in the LSAI. It is expected that the more household size have more employee member in LSAI than the less household size.

4. Household education level- it is a dummy variable with value of 1 for those who are literate (who are attending formal school), 0 otherwise. It is expected that the more educated household get new technologies from agricultural investment and transfer to investors than the illiterate one.

a) Occupation of household- It is a dummy variable with value of 1 if the household headed employed in the farming activities, other wise 0. It is expected that household which participate in farm activities have more approach to LSAI and can get new technologies and employment opportunity than others.

b) Large scale agricultural investment:- The total number of agricultural investment projects having land owned.

c) Technology transfer:- it is a dummy variable with the value of 1 for those household who get technology from LSAI and 0 other wise.

d) Infrastructure developed:- The infrastructure developed due to the expansion of LSAI.

e) Loss of land:- it’s the dummy variable with the value 1 for the household loss the land due to expansion of LSAI and 0 other wise.

f) Distance of HH residence from the investment project: - it is a dummy variable with the value of 1 for the HH residence nearest to the agricultural investment and 0 if the HH residence far from the agricultural investment. It is expected that the HH nearest to the agricultural investment be affected by any factors than the HH with far from the agricultural investment.

g) Dependent Variables.

h) Employment (as treatment and control):- It is a dummy variable with 1 for household having one or more employee members in LSAI and 0 other wise. Treatment group for household having one or more employee members in the large-scale agricultural investment and Control group for the household do not have employee members in large-scale agricultural investment.

i) Outcome Variables

j) Household income: - The household incomes generated are from crop production, livestock farming, forest products, irrigation and other off-farm incomes. It’s expected that the expansion of LSAI will increase the income at the household level through poverty reduction, employment opportunity creation, productivity improved, production technology and other related activities.

k) Household asset growth: - The asset growth due to expansion of LSAI at the household level will be the land holding, residence (dwelling), machinery (farm machinery), livestock production, financial assets and other resources accumulated by the household. It is expected that the expansion of LSAI project will increase the household asset accumulation through the transfer of new technology, improve working culture, productivity improvement, availability of infrastructure and others related activities.

3. Result and Discussions

3.1. Descriptive Analysis

It was hypothesized that the social, economic and institutional features of the households—such as age of HH, sex of HH, educational attainment, size of the household, technology transfer, type of occupation, developed infrastructure, loss of land, and distance from investment projects—would have an effect on the communities involved in the large-scale agricultural program, which would then have an impact on the outcomes variables, like household assets and income. Crop production, livestock products and supplies, off-farm revenue, revenue from forest goods, and revenue from irrigations are the sources of income for the households. A household's assets include its land, house, vehicle, animals, produce, and money. Out of the 330 sample respondents in total, 160 were treated household members who received employment opportunities from investment projects, and the remaining 170 were control households, located far away from investment projects, where their HH members did not receive employment opportunities from investment project. The (table 4) below shown has the following Summery statistics tables for categorical and dummy variables.

Table 4. Summery Statistics categorical variables.

Variable	Obs	Categorical variable	frequency	Percent
marital status of Households	330	Married	252	76.36
		Never married	49	14.8
		Widow	29	8.79
Information about investment owning land	330	kebele leaders"	132	40.0
		Kebele land management committee	40	26.7
		government officials	107	32.4
		Investors	3	0.9
Investment affected household income	330	loss of farm land	33	10
		absences of job opportunity	7	2.1
		production decrease	2	0.6
		income generating forest and deforest	178	53.9
		All	110	33.3
Type of land losses	330	crop land	34	10.3
		grazing land	48	14.5
		grass land	1	.3
		source of forest products	50	15.2
		All	1	.3
		no loss of land	196	59.4
Agreement with the transparency of land deals	330	strongly agree	8	2.4
		Agree	100	30.3
		Neutral	66	20.0
		Disagree	119	36.1
		strongly disagree	37	11.2
Extent of the direct effects of investment projects on means of living	330	High	78	23.6
		Medium	222	67.3
		low"	30	9.1
Investment project investing in your area affected you	330	security of land holding	14	4.2
		access to crop land	8	2.4
		access to grazing land	45	13.6
		access to forest land and forest products	223	67.6
		access of water for drinking	34	10.3
		access of water for your animals	3	.9
		access of water for irrigation	1	.3
		None	2	.6
HH livelihoods option & opportunity affected by investment projects	330	loss of forest products do to deforestation	8	2.4
		crop production and productivity decrease	3	.9
		computation on grazing land	21	6.4
		farm land grabbing	39	11.8
		All	259	78.5
Opportunity that investment project provide	330	employment opportunity creation	25	7.6
		technology transfer	23	7.0
		utilization of agricultural inputs increase	12	3.6
		productivity of crop increase	66	20.0
		working culture of the community change	6	1.8
		None	198	60.0
Livelihoods of household change	330	asset accumulation of household improved	81	24.5
		food security problems of the household improved	1	.3
		employment opportunity generated	99	30.0
		None	149	45.2
Variable	Dummy
Sex of household	330	Female	37	11.2
Sex of household	330	Male	293	88.8
Educational levels of household	330	"literate"	218	66.1
Educational levels of household	330	illiterate	112	33.9
Distances of household residence from investment projects	330	"far from investment projects"	119	36.1
Distances of household residence from investment projects	330	"nearest to investment projects"	211	63.9
Rate of poverty	330	"increase"	36	10.9
Rate of poverty	330	"decrease"	294	89.1
Technology get from investment project	330	"no"	129	39.1
Technology get from investment project	330	"yes"	201	60.9
Employment opportunity from investment Projects	330	treated	160	48.5
Employment opportunity from investment Projects	330	control	170	51.5
Occupations of the household	330	"farming "	232	70.3
Occupations of the household	330	"both farming and trade"	98	29.7
Loss of useful land due to investment project	330	"no"	114	34.5
Loss of useful land due to investment project	330	"yes"	216	65.5
Infrastructure develop by investment projects	330	yes	70	21.2
Infrastructure develop by investment projects	330	No	260	78.8
Large-scale agricultural investments has contribution on household poverty reduction	330	yes	177	53.6
	330	No	153	46.4
Consultation when land transferred to Investors	330	yes	106	32.1
Consultation when land transferred to Investors	330	No	224	67.9
Evicted from home because of investment Projects	330	high	20	6.1
		medium	231	70.0
		low	79	23.9
HH face food shortage last 12 months	330	yes	276	83.6
HH face food shortage last 12 months	330	No	54	16.4

Source: - own Survey of 2023

Table 5. Summery Statistics continues variables.

Variable	Obs	Mean	Std. Dev.	Min	Max
Age of household	330	40.48	9.81054	22	66
Size of the household	330	6.05	3.512	1	27
Numbers of employment opportunity created Permanent	330	.0697674	.2977773	0	2
Numbers of employment opportunity created Temporary	330	.6104651	1.258657	0	8
Size of the land lost because of land investment	330	1.431	2.65435	0	10
Months of households food production for their own use	330	2.610465	.9012699	1	4

3.2. Characteristics of Continuous Variables

The descriptive results of continues variables for the whole sample of the households those nearest to investment projects affected by the LSAI (treatment group) and those far from investment projects not more affected by LSAI (control group). The mean difference test between the treatment group and the control groups are presented in the table 5 above.

3.3. Mean of Continuous Variables Characteristics of Respondent

Table 6's descriptive data demonstrate that there is no discernible difference in the employment status of households based on age. However, the mean responder size differs statistically significantly between respondents who are employed and those who are not. The total sample's average land loss as a result of investment projects is 2.43, while the treatment groups' average land loss (households closest to the investment and farmlands) is 1.434 and the control groups' average land loss (far from investment projects no any loss of lands) is 1.427. The size of land lost as a result of investment projects did not significantly differ, according to the results. In terms of size of land lost due to investment projects, result indicated there is no significant difference on loss land.

The results of the Pearson's chi square proportions test for dummy variables across the treatment and control groups show that the LSAI affects local populations' livelihoods either directly or indirectly (Table 7). Consequently, 216 respondents (or 65%) of the selected population affirm that there has been a loss of use of land as a result of investment projects, whereas 114 respondents (or 0.34%) stated that there has been no loss of use of all land. It can be inferred from this that households closest to investment projects lose more land than ones farther away. There was inadequate community consultation done throughout the land transfer to investment.

Based on the findings presented in Table 6, 224 (68%) of the sample respondents stated that no community consultation was carried out prior to the land transfer for investment. 106 respondents, or 32%, confirmed that community consultations were held prior to the transfer of land for investment. This indicates that no consultation will take place before to the land being handed to the investment. Because of the investment projects' expansion, the households closest to them are forcibly removed from their homes. According to the study, 37 treatment groups (or 23.26%) out of all the groups have had some level of home displacement, whereas the control groups have experienced none.

When a household is forced to leave their home owing to settlement in the region, their land is fully turned over to investors. According to the poll results, 6% of the control group and 20% of the treatment group said that farmers are being forced to leave their homes because the project is growing. Furthermore, 94% of the control group and 8% of the treatment group concur that there has never been a farmer evicted as a result of an investment project. The chi square value indicated that there is no statistically significant difference between the treated and control groups when it comes to house evictions brought on by investment projects.

Table 6. Descriptive statistics and mean different test between the continuous variables.

Variables	Employed		Unemployed		Total		t- value
Variables	Mean	Std	Mean	Std	Mean	Std	t- value
Age of HH	41.225	10.33	39.788	9.60	40.54	9.83	1.3084
Size of HH	3.6	1.939	8.34	3.09	6.04	3.51	-16.5233 ***
Size of land lost because of land Investment	2.434	2.75414	1.427	2.565047	1.431	.146117	0.0220
Numbers of employee opportunity created permanently	0.13	0.04	0.01	0.01	0.07	0.03	-5.56***

Source: - Own survey2023

Note: - *, **, *** significant at 10%, 5% and 1% probability level

According to the study, 30 members of the treatment group and 13% of the control group out of the total sampled respondents said that the project offers opportunities for employment, technology transfer, the use of agricultural inputs, altering the community's working culture, and productivity. On the other hand, roughly 3% of the treatment group and 9% of the control group thought the investment project offered no opportunities for the community. The statistical significance between the treated and control groups for opportunity investment proved for the household and communities are indicated by the chi square value.

In terms of technology transfer from the investment projects, Out of all responders, 38 (44.19%) of the treatment groups receive different technologies from the investment projects than do the 12 (13.95%) control groups. The chi square result indicates that the HH receives technology from an investment project with statistical relevance. As the study showed, LSAI makes a very small contribution to reducing poverty; among the sampled respondents, 81% of the treatment group and 95% of the control group stated that the project did not lower the level of poverty, with the remaining 18% of treatment groups and 4% of the control group saying the same. The rate of poverty is rising sharply, meaning that 21% of those receiving treatment and control report that the increase in poverty is a result of increased agricultural investment in the region.

According to the chi square result, there is no statistically significant difference in the rate of poverty between the treatment and control groups. Within a year of crop production or within a year of facing a food crisis, 73% of the total respondents indicated that they had experienced a food shortfall in the previous year. The findings indicate that households closest to investment projects have a greater food shortage than households farther away from investment projects. Expansions of investment ventures into new areas can help the community by constructing new infrastructure. As a result, 21% of respondents from the treatment and control groups state that the research area lacks any infrastructure created by investment projects.

Table 7. Descriptive statistics and percentage difference test for the dummy variables.

Dummy variables	Category	Treatment		Control group		Total group
Dummy variables	Category	N	%	N	%	N	%	x²
HH Loss of valuable land due to investment projects	Yes	49	.93	167	0.98	216	.65	0.6410***
HH Loss of valuable land due to investment projects	No	111	0.3	3	0.017	114	0.345	0.6410***
Consultation in the course of land transferred to investment	Yes	48	0.3	58	0.34	106	0.32	0.6410
	No	112	0.7	112	0.7	224	0.68	0.6410
Evicted from home due to investment projects	Yes	37	0.2	11	0.06	48	0.15	30.661
Evicted from home due to investment projects	No	122	0.8	160	0.94	282	0.85	30.661
Opportunity investment provided for the household and community	Yes	46	0.3	22	0.13	68	0.2	4.65
	No	114	0.3	148	0.9	262	0.8	4.65
Technology that the household acquire from land investment project	Yes	44	0.28	157	0.92	201	0.6	145.59
	No	116	0.725	13	0.08	129	0.39	145.59
Rate of poverty	Yes	29	0.18	7	0.04	36	0.11	16.639
Rate of poverty	No	131	0.818	163	0,95	294	0.8	16.639
LSAI support to HH poverty reduction	yes	81	0.5	96	0.56	177	0.54	0.287
LSAI support to HH poverty reduction	no	79	0.49	74	0.43	153	0.46	0.287
Infrastructure develop by investment projects	yes	34	0.21	36	0.21	70	0.22	0.0003
Infrastructure develop by investment projects	no	126	0.78	134	0.78	260	0.78	0.0003
HH face food deficiency last 12 months	yes	134	0.83	142	0.83	276	0.73	0.0029
HH face food deficiency last 12 months	no	26	0.16	28	0.17	54	0.16	0.0029

Source;-Own survey data of 2023

3.4. Mean Diverse Test of the Outcome Variable

The household's income and asset accumulation represent the mean different test of the study's outcome variables. The land holding capacity of the household, dwelling (home), machinery (equipment), livestock, fruits, and other financial assets account for the entirety of the asset accumulations. Crop output, livestock products, off-farm income, forest goods, and irrigation revenue account for the households' overall income.

Table 8. Mean diverse test for the outcome variable.

Variables	Unit	Treatment group		Control group		t-value
Variables	Unit	Mean	Std	Mean	Std
Total asset growth of the HH	Birr	1.63125	.3387338	1.411765	.4936069	-5.9835***
Total revenue of the HH	Birr	1.506	.4596964	1.3	1.429961	-3.8807***

The average total asset buildup of the household is 1.63125 for the treatment groups closest to the investment projects in terms of residence and farmland, and 1.411 for the control groups farther away from the projects in terms of residence and farmland, representing asset values greater than 0-100,000 birr. The treatment groups' total household income is 1.506, signifying an increase in income over 30,000 birr, whereas the control groups' total income is 1.3, signifying an increase in income below 30,000 birr. The findings indicate that there is a statistically significant difference at 1% for income and total asset accumulation, respectively.

3.5. Econometric Analysis

In order to increase the quality estimate, balancing tests and certain indicators were used to determine the suitability of Propensity Score Matching as the estimator. The econometric analysis is carried out to determine the factors that influence employment opportunities, technology transfer, infrastructure development, changes in household livelihoods, land loss and the amount of land taken up by investment projects, opportunities that investment projects offer for the local community, the contribution of LSAI to the reduction of poverty, the rate of poverty, the opportunities and livelihood options available to households, the impact of investment projects on household income and asset accumulation, and more.

3.5.1. Logit Model Determinants of LSAI on Livelihoods of Local Community

A binary logit regression model was used to describe how large-scale agriculture affected the local community's standard of living. The findings indicate that only six variables were found to be statistically significant out of the characteristics that were expected to influence local community employment in large-scale agricultural land investment projects (LSAI). These include of the household's educational attainment, size, occupation, and proximity to investment projects, among other things. The employment opportunity projects are greatly and favorably impacted by the loss of land resulting from investment and technology transfer, whereas the distance of a household's home from an investment project has a negative impact (Table 9). A household's proximity to a large-scale agricultural investment has an impact on employment and community involvement in the area.

Table 9. Binary logit regression of households’ employment opportunity from LSAI.

Employment opportunity investment project	Coef.	Std. Err.	Z	P>\|z\|
Sex of the household head	7.106623	17.9521	0.78	0.438
Age of the household head	.9779019	.0445749	-1.01	0.701
Educational levels of household	98.88295	130.5764	3.48	0.001
Occupation of the household	4.51284	3.837754	1.77	0.076
Size of household	2.119105	.3389252	4.70	0.000
Remoteness of HH residence from land investment project	23.21895	25.55457	2.86	0.004
Loss of valuable land due to land investment Project	895.96	1454.46	4.19	0.000
Size of land lost for the reason that of land investment	0.329943	.4384427	0.86	0.387
Rate of poverty	.2570666	.4101169	-0.85	0.395
Technology get from investment Project	185.8994	244.0064	3.98	0.000
Infrastructure developed by investment projects	4.65673	4.952327	0.148	0.148
Cons	9.99e-12	5.82e-11	-4.35	0.000
Marginal effect after logit model
Variable name
Sex of the household head	.3721002	.28417	0.31	0.438
Age of the household head	-.005499	.01117	-0.51	0.599
Learning levels of household head	1.130496	.34109	3.31	0.001***
Occupation of the household	.3708311	.20943	1.77	0.077**
Size of household	.1848078	.04033	4.58	0.000***
Remoteness of HH head residence from investment project	.6130576	.1461	4.20	0.000***
Loss of valuable land due to investment Project	.8815833	.07672	11.49	0.000***
Size of land lost for the reason that of land investment	.0701674	.08299	0.85	0.387
Degree of poverty	-.3214855	.3279	-0.98	0.327
Technology acquired from investment Project	.8258028	.09192	8.98	0.000***
Infrastructure developed by investment projects	.3785554	.2657	1.42	0.154

Source: - Own survey data of 2023

Note: - *, **, *** significant at 10% and 5%, and 1% probability level

3.5.2. Matching Estimates of the Propensity Score

In collective action schemes, the histograms of estimated PS density for treatment groups and control groups overlap the common provision section graphs, as seen in Figure 1. Participants who receive appropriate support are indicated by the term "treated on supports," whereas those who receive inappropriate support are indicated by the term "treated off supports."

The graphs showed that, with the exception of some treated off support individuals; all treated and untreated persons were located inside the zones of common provision, suggesting that every treated individual had a matching untreated individual. Although some treated off-support individuals will not be comparable to the untreated persons, this ensures that statistically treated and untreated individuals are equivalent. This guarantees statistical parity between persons who have received treatment and those who have not; nonetheless, certain treated off-support patients will not be comparable to the untreated individuals.

Download: Download full-size image

Figure 1. Propensity score spreading and common provision for propensity score estimation.

3.5.3. Matching Estimation Procedures

Propensity score results for the effects of LSAI on local community livelihoods were estimated with the aim of determining whether our cross-sectional matching estimators are sensitive to sample size choices, common support conditions that are enforced, and whether balancing propensity is set and satisfied in all regressions at the 1% significant level. The probability of the effects that the LSAI addresses on the daily activities of the local community is predicted using the logit model, which also takes into account a variety of household aspects. Table 10 presents radius matching, kernel matching, and nearest neighbor matching which are helpful tools for ensuring that the estimated causal effects are consistent.

Table 10. Performances of matching estimators.

Matching algorithm	Psedo-R2	Insignificant Variables	Sample size matched
Nearest Neighbor matching (NNM)
NNM (1)	0.528	10/11	206
NNM (2)	0.540	8/11	207
NNM (3)	0.337	5/11	207
NNM (4)	0.298	8/11	207
NNM (5)	0.265	9/11	207
Caliper match (CM)
Caliper (0.01)	1.000	7/11	182
Caliper (0.1)	0.528	10/11	206
Caliper (0.25)	0.528	10/11	55
Caliper (0.5)	0.540	8/11	207
Radius match (RM)
Radius (0.01)	0.851	3/11	207
Radius (0.1)	0.851	3/11	207
Radius (0.25)	0.851	3/11	207
Radius (0.5)	0.851	3/11	207
Kernel matching (KM)
Kernel (0.01)	1.00	9/11	182
Kernel (0.1)	1.00	11/11	207
Kernel (0.25)	1.00	9/11	207
Kernel (0.5)	1.00	11/11	207

The number of matched observations, the pseudo R-square value, and the insignificant variables are the three criteria used to choose the best matching algorithms through matching the various algorithms for the estimations of the treatment effects. The optimal matching algorithm, as determined by the selection criteria, is kernel matching (KM) with a band width of 0.1.

3.5.4. Balancing Tests

According to the t-test balancing tests, there are some household characteristic differences between the treatment and control groups that are significant in other ways and jointly insignificant before and after matching. The variables that do not have statistically significant mean differences among the numbers in the individual covariates balances tests are the following: sample size matchless (sex of HH, age of HH, educational levels of HH, occupation of the HH, and size of the HH); sample KM at bandwidth with 0.25 (all variables haven’t significant differences); sample after KM (0.1) bandwidth (all variables haven’t significant differences); and sample after caliper radius with bandwidth at 0.25 (sex of HH, age of HH, HH educational levels, occupation of the HH and size of the HH have no significant differences). At 10%, 5%, and 1%, the remaining differences were statistically significant. After matching processes, the balancing test demonstrates that the numbers of covariates stay balanced.

Stated differently, following the matching methods, there is no discernible difference between the treatment and control groups' covariate means and frequency distributions. As a result, the treatment effects estimation results are expressed and implemented using kernel matching. We perform the ATT estimate using the best choice estimator, which is the kernel matching band width (0.1) matching approaches, as determined by the study. Based on all of the experiments, it appears that the selected matching algorithm performs relatively better with the available data. As a result, we are able to estimate the households' average treatment effects (ATT).

Table 11. Propensity scores and Covariate balances.

Variable	Sample	Treated	Controls	Difference	Bias (%)	T test
Propensity score	Un matched	0.41021	0.37765	0.03256	15.0	0.32
Propensity score	Matched	0.42033	0.41932	0.00101	15.0	0.02
Sex	Un matched	0.87	0.79	0.08	9.4	1.15*
Sex	Matched	0.84	0.88	-0.04	9.4	-0.45
Age	Un matched	40.26	40.62	-0.37	-7.2	-0.21
Age	Matched	40.91	41.84	-0.93	-7.2	-0.44
Education levels	Un matched	0.54	0.61	-0.07	-11.4	-0.79
Education levels	Matched	0.53	0.54	-0.01	-11.4	-0.11
Occupation	Un matched	0.87	0.87	0	1.3	-0.01
Occupation	Matched	0.84	0.88	-0.04	1.3	-0.5
Household Size	Un matched	9.08	8.48	0.6	1.4	0.8
Household Size	Matched	9.31	10.11	-0.8	1.4	-0.76
Distances from investment project	Un matched	0.67	0.45	0.22	19.7	2.39**
Distances from investment project	Matched	0.63	0.61	0.02		0.13
Loss of land to investment projects	Un matched	0.28	0.17	0.12	9.7	1.63*
Loss of land to investment projects	Matched	0.22	0.28	-0.06	9.7	-0.65
Size of land taken for investment	Un matched	1.87	0.64	1.23	15.7	2.57**
Size of land taken for investment	Matched	1.34	1.68	-0.33	15.7	-0.48
Rate of poverty	Un matched	-0.48	0.03	0.2	42.2	4.37**
Rate of poverty	Matched	0.16	0.13	0.03	42.2	0.44
Technology transfer	Un matched	0.62	0.2	0.42	45.7	5.48***
Technology transfer	Matched	0.53	0.52	0.02	45.7	0.14
Infrastructure developed	Un matched	0.08	0.01	0.07	15.3	2.56**
Infrastructure developed	Matched	0.03	0.02	0.01	15.3	0.39

Note: *, **, *** significant at 10%, 5% and 1% probability levels

Source: own data estimation results

3.5.5. Treatment Effects on the Treated (ATT)

The study finding provides the evidences as to whether or not an effects of LSAI on the livelihoods of the local communities have bring important change on the total household asset growth and total income of the household.

Table 12. Average treatment effects on treated (ATT).

Variable	Treated	Controls	Difference	S.E.	T-stat
Total asset growth of household head	1.15625	1.28458361	-.128333607	.086586478	-1.48*
Total income of household head	1.34375	1.45814267	-.114392666	.11026736	-1.04*

Sources: Own survey results; * significant at 10% probability levels

When the estimation results were compared to the total household income in birr, which is gained from various livelihood options such as crop production, livestock products, forest products, irrigation, and off-farm incomes, they provided supportive evidence of statistically significant differences in the household's total asset accumulations.

The total asset buildups of households and the total household revenue by 10% probability levels differ statistically significant, according to the propensity score data. The findings indicate a negative correlation between household income and asset accumulation resulting from large-scale agricultural investments.

3.6. The Effect of Land Transfer to Large Scale Agricultural Land Investment Projects on Local Communities

Large-scale agricultural land transfers for investments have both beneficial and negative consequences on the standard of living in the surrounding populations. According to the main interviews, the Woreda administration provided more information and made more choices regarding the transfer of land for investment than did the local communities. 40 percent (132) of the information regarding the land transfer to the communities came from government members, 32.4% (107), Keble leaders, 26.6% (88), the land management committee, and 1% (3) from investors. This suggests that land has been transferred to large-scale agricultural land investment without communities’ consultation or agreement.

Accordingly, there is roughly 36.1% disagreement (119 people), 20% neutrality (66 people), 15.1% agreement (50 people), 26.4% strong disagreement (87 people), and 2.4% strong agreement (8 people). The majority of households do not find large-scale agricultural land investments appealing due to the greater number of negative effects than favorable ones, according to the statistics.

According to the survey, 10.3% (44) of the land lost as a result of large-scale agricultural land investment projects was lost as crop land, 15.5% (50) as a source of forest products, 14.5 (48) as grazing land, and the remaining 59.4% (196) did not lose any valuable land.

Thus, the local inhabitants' means of subsistence are directly impacted when land is transferred for extensive agricultural investment. According to the results, over 76.74 percent of respondents depend on forests and forest products for their livelihoods.

3.7. Opportunity of Large Scale Agricultural Land Investment for the Local community

For the local communities, the growth of large-scale agricultural investment presents new opportunities. The possibility that investment projects offer to the nearby communities was another major focus of the study. Of the 330 respondents, 7.6%(25) said that no opportunities would be created for the local community; 7.0%(23) said that technology would be transferred; 1.8% said that the community's working culture would change; 3.6%(12) said that the use of agricultural inputs would increase; 20.0%(66) said that crop productivity would increase; and 60% (198) said that no opportunities would be created for the local community. The aforementioned finding suggests that large-scale agricultural investment initiatives did not offer any opportunities for the local people.

In response to the proposal to modify LSAI to improve household income and the lives of the local community, 10.47 percent (36 respondents) rated the creation of job opportunities as positively impactful, 1.16 percent (4) said that the household's asset growth improved, 0.58% said that the household's problems with food security were resolved, and 86.79% said that their livelihoods would remain the same. The findings show that there is no discernible impact of large-scale agricultural land expansion on the local communities’ standard of living within the study region.

4. Conclusion

The aim of the study was to investigate effects of large scale agricultural land investments on the livelihoods of communities. The whole of the land rented to investment projects and the cultured areas are not being used in accordance with community agreements and intentions. Few of the overall number of investors in the area make excellent returns; the majority of investment projects lack sufficient equipment, the sites are not automated, opportunities for the communities were not produced, and the land is unlawfully rented to other farmers. The land transfer procedures are opaque; communities are not consulted; instead, the land transfer is approved by the woreda investment committee and is only discussed with the Kebele administration and land use committees. This keeps the land transfer hidden from the community. A community's opportunity to support itself was lost as a result of the unseen land transfer to investors.

The local communities and investors have very little interaction; investors deal with Kebele administration when issues arise and they need labor on a daily basis for cropping, weeding, and harvesting, unless the local community is not involved as a permanent employee. In comparison to the quantity of investment projects, the opportunities generated for the local populations by large-scale agricultural investments are quite modest. Descriptive statistics and econometric analysis were used to assess the data, with comparisons made between treatment groups (those closest to investment projects) and control groups (those farthest from investment projects). The household asset accumulations are the outcome variables impacted by large-scale agriculture.

Descriptive analysis findings indicate that there is a statistically significant difference between the treatment and control groups.

Regression analysis results demonstrate that the explanatory variables—household sex, proximity to investment projects, land loss as a result of investment project, degrees of poverty, and technology transfer—have a positive impact on households' employment opportunities in large-scale agricultural investments. The average treatment effects estimation results on the treated show that large-scale agriculture land investments has detrimental effects on household incomes and asset buildups.

The study's conclusions led to the following recommendations regarding drowning:

1) The lack of openness in the land transfer to agricultural investment can be attributed to the lack of decision-making and community consultation. Therefore, while transferring land for agricultural projects, the government pays attention and the community must be consulted.

2) The management and implementation of land transfer for large-scale agricultural land investment projects is inadequate. Consequently, the governments oversee and manage the land that is transferred for agricultural land investment and regularly assess the level of performance.

3) The socio-economic and ecological effects of a large-scale agricultural investment must be carefully considered before transferring the land for investment.

4) Alternatives to agriculture are necessary for the local community's livelihoods in order to survive. There were no longer any medicinal plants or trees that produced honey in the areas' natural forest, which had been cleared of snow and destroyed. Consequently, when land is transferred to large-scale agricultural investments, the government must compromise natural forest degradation.

5) Investment initiatives in the research regions deplete natural resources, even though they create opportunity for the local people. The loss of resources as outcome of the growth of large-scale agricultural land investment is something that the governments are aware of.

Abbreviations

BGRS: Benishangul Gumuz Regional State

BGRSIO: Benishangul Gumuz Regional Sate Investment Office

BWARD: Bambasi Woreda Agriculture and Rural Development Office

HH: Households

CAADP: Comprehensive Africa Agriculture Development Program

FAO: Food and Agriculture Organization

FDRE: Federal Democratic Republic Of Ethiopia

FGD: Focus Group Discussion

GDP: Growth Domestic Product

LSAI/P: Large-Scale Agricultural Investment Project

MOARD: Minister of Agriculture and Rural Development

Author Contributions

Shafe Zelalem Gasisa: Conceptualization, Data curation, Formal Analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

Teha Romanu Benti: Conceptualization, Data curation, Formal Analysis, Funding acquisition, Investigation, Methodology, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

Mihret Fentahun Yeneneh: Formal Analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – review & editing

Conflicts of Interest

The authors declare no conflicts of interest.

References

[1]	K. Nolte, W. Chamberlain, and M. Giger, International Land Deals for Agriculture: fresh insights from the Land Matrix: Analytical Report II. 2016. https://doi.org/10.7892/boris.85304
[2]	D. Teklemariam, H. Azadi, J. Nyssen, M. Haile, and F. Witlox, “How sustainable is transnational farmland acquisition in Ethiopia? Lessons learned from the Benishangul-Gumuz Region,” Sustain., vol. 8, no. 3, pp. 1–27, 2016, https://doi.org/10.3390/su8030213
[3]	S. Kolavalli, R. Birner, and K. Flaherty, “The Comprehensive Africa Agriculture Program as a Collective Institution,” SSRN Electron. J., no. December, 2013, https://doi.org/10.2139/ssrn.2197400
[4]	W. M. Azeb W. Degife, “Socio-economic and Environmental Impacts of Large-Scale Agricultural Investment in Gambella Region, Ethiopia,” J. US-China Public Adm., vol. 14, no. 4, pp. 183–197, 2017, https://doi.org/10.17265/1548-6591/2017.04.001
[5]	D. K. Ketema, B. Emanna, and G. Tesfay, “Impact of land acquisition for large-scale agricultural investments on vulnerability of displaced households to climate change shocks in Ethiopia,” Ecosyst. People, vol. 18, no. 1, pp. 643–660, 2022, https://doi.org/10.1080/26395916.2022.2143572
[6]	A. K. Guyalo, E. A. Alemu, and D. T. Degaga, “Impact of large-scale agricultural investments on the food security status of local community in Gambella region, Ethiopia,” Agric. Food Secur., vol. 11, no. 1, pp. 1–28, 2022, https://doi.org/10.1186/s40066-022-00381-6
[7]	FAO, “AQUASTAT Country Profile – Ethiopia. Food and Agriculture Organization of the United Nations (FAO). Rome, Italy,” FAO, AQUSAT reports, pp. 11–12, 2016.
[8]	MoARD, “Federal Democratic Republic of Ethiopia Ministry of Agriculture and Rural Development,” Minist. Agric. Rural Dev. Ethiop. Agric. Sect. Policy Invest. Framew., vol. 2010, no. June, pp. 2009–2012, 2010.
[9]	World Bank, “UNLOCKING AFRICA ’ S AGRICULTURAL Unlocking Africa ’ s Agricultural Potential,” 2013.
[10]	T. Moreda, “Large-scale land acquisitions, state authority and indigenous local communities: insights from Ethiopia,” Third World Q., vol. 38, no. 3, pp. 698–716, 2017, https://doi.org/10.1080/01436597.2016.1191941
[11]	G. Alemu, “Rural Land Policy, Rural Transformation and Recent Trends in Large-Scale Rural Acquisitions in Ethiopia,” p. 28, 2011.
[12]	R. Dessalegn, “Land to investors: Large-Scale Land Transfers in Ethiopia,” L. Gov. equitable Sustain. Dev., pp. 0–36, 2011.
[13]	M. S. Bekele, “Economic and Agricultural Transformation through Large-scale Farming Impacts of large-scale farming on local economic development,” no. October 2016, pp. 1–288, 2016.
[14]	G. A. Alamineh and K. S. Anteneh, “DEVELOPMENTAL PARADOX IN ETHIOPIA: LARGE SCALE AGRICULTURE AND ITS IMPACT,” J. Wind Eng. Ind. Aerodyn., vol. 26, no. 1, pp. 1–4, 2019.
[15]	G. D. Isreal, “Using Published Tables Using Formulas To Calculate A Sample Size Using A Census For Small Populations”.

Cite This Article

Plain Text BibTeX RIS

APA Style

Gasisa, S. Z., Yeneneh, M. F., Benti, T. R. (2024). Effects of Large-Scale Agricultural Land Investment on Local Communities Livelihoods: Evidence from Bambasi Woreda, Western Ethiopia. American Journal of Environmental and Resource Economics, 9(2), 20-39. https://doi.org/10.11648/j.ajere.20240902.11

Copy | Download

ACS Style

Gasisa, S. Z.; Yeneneh, M. F.; Benti, T. R. Effects of Large-Scale Agricultural Land Investment on Local Communities Livelihoods: Evidence from Bambasi Woreda, Western Ethiopia. Am. J. Environ. Resour. Econ. 2024, 9(2), 20-39. doi: 10.11648/j.ajere.20240902.11

Copy | Download

AMA Style

Gasisa SZ, Yeneneh MF, Benti TR. Effects of Large-Scale Agricultural Land Investment on Local Communities Livelihoods: Evidence from Bambasi Woreda, Western Ethiopia. Am J Environ Resour Econ. 2024;9(2):20-39. doi: 10.11648/j.ajere.20240902.11

Copy | Download

@article{10.11648/j.ajere.20240902.11,
  author = {Shafe Zelalem Gasisa and Mihret Fentahun Yeneneh and Teha Romanu Benti},
  title = {Effects of Large-Scale Agricultural Land Investment on Local Communities Livelihoods: Evidence from Bambasi Woreda, Western Ethiopia
},
  journal = {American Journal of Environmental and Resource Economics},
  volume = {9},
  number = {2},
  pages = {20-39},
  doi = {10.11648/j.ajere.20240902.11},
  url = {https://doi.org/10.11648/j.ajere.20240902.11},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ajere.20240902.11},
  abstract = {Most of the African countries including Ethiopia are often known by problem of large scale agricultural land investment effects on the local community where land is being transferred to investment. Even though several efforts made so far to solve the overall effects of large scale agricultural land investment situation, the challenge is still widespread problem in Ethiopia. Hence the study project's purpose was to find out how large agricultural investments in Bambasi Woreda, Western Ethiopia impact the livelihoods of the surrounding populations and examine the local communities' participation in the large scale agricultural investment. In order to attain these objectives, data were collected from 330 randomly selected households in four purposively selected kebeles of the district for both control groups and treatment groups. The sample size was chosen using a multistage stratified random sampling technique. Both qualitative and quantitative data gathering techniques and instruments were employed in the study. Besides, the instruments utilized to collect the data were observations, focus groups, interviews, household surveys, and document reviews. Data was analyzed using both descriptive statistics and econometric methods. The study shows that out of the total sampled respondents 30 of the treatment group and 13 percent of control group reveal that the project provided opportunity in terms of employment opportunity, technology transfer, utilization of agricultural inputs, changing the working culture of the community and productivity. The chie square value shows there is statistical significance among treated and control group on opportunity investment provided for the household and community. A binary logit regression model was used to describe how large-scale agriculture land investment affected the local community's standard of living. The findings indicated that only six variables were found to be significant out of the characteristics that were expected to influence local community employment in large-scale agricultural land investment projects. These includes household's educational accomplishment, size of HH, occupation of HH, Loss of useful land due to investment Project and technology transfer significantly and positively affected the employment opportunity in large scale agricultural investment projects, whereas the distance of a household's home from an investment project has a negative impact. Large-scale agricultural investments have a detrimental influence on household wealth accumulation and income, according to the estimation results of the average treatment effects on the treated. The management and implementation of land transfer for large-scale agricultural investment projects is inadequate, lack of openness, absence of community consultation, natural forest degradation, socio-economic and ecological effects must be carefully considered before transferring the land for large-scale agricultural investment.
},
 year = {2024}
}

Copy | Download

TY  - JOUR
T1  - Effects of Large-Scale Agricultural Land Investment on Local Communities Livelihoods: Evidence from Bambasi Woreda, Western Ethiopia

AU  - Shafe Zelalem Gasisa
AU  - Mihret Fentahun Yeneneh
AU  - Teha Romanu Benti
Y1  - 2024/05/10
PY  - 2024
N1  - https://doi.org/10.11648/j.ajere.20240902.11
DO  - 10.11648/j.ajere.20240902.11
T2  - American Journal of Environmental and Resource Economics
JF  - American Journal of Environmental and Resource Economics
JO  - American Journal of Environmental and Resource Economics
SP  - 20
EP  - 39
PB  - Science Publishing Group
SN  - 2578-787X
UR  - https://doi.org/10.11648/j.ajere.20240902.11
AB  - Most of the African countries including Ethiopia are often known by problem of large scale agricultural land investment effects on the local community where land is being transferred to investment. Even though several efforts made so far to solve the overall effects of large scale agricultural land investment situation, the challenge is still widespread problem in Ethiopia. Hence the study project's purpose was to find out how large agricultural investments in Bambasi Woreda, Western Ethiopia impact the livelihoods of the surrounding populations and examine the local communities' participation in the large scale agricultural investment. In order to attain these objectives, data were collected from 330 randomly selected households in four purposively selected kebeles of the district for both control groups and treatment groups. The sample size was chosen using a multistage stratified random sampling technique. Both qualitative and quantitative data gathering techniques and instruments were employed in the study. Besides, the instruments utilized to collect the data were observations, focus groups, interviews, household surveys, and document reviews. Data was analyzed using both descriptive statistics and econometric methods. The study shows that out of the total sampled respondents 30 of the treatment group and 13 percent of control group reveal that the project provided opportunity in terms of employment opportunity, technology transfer, utilization of agricultural inputs, changing the working culture of the community and productivity. The chie square value shows there is statistical significance among treated and control group on opportunity investment provided for the household and community. A binary logit regression model was used to describe how large-scale agriculture land investment affected the local community's standard of living. The findings indicated that only six variables were found to be significant out of the characteristics that were expected to influence local community employment in large-scale agricultural land investment projects. These includes household's educational accomplishment, size of HH, occupation of HH, Loss of useful land due to investment Project and technology transfer significantly and positively affected the employment opportunity in large scale agricultural investment projects, whereas the distance of a household's home from an investment project has a negative impact. Large-scale agricultural investments have a detrimental influence on household wealth accumulation and income, according to the estimation results of the average treatment effects on the treated. The management and implementation of land transfer for large-scale agricultural investment projects is inadequate, lack of openness, absence of community consultation, natural forest degradation, socio-economic and ecological effects must be carefully considered before transferring the land for large-scale agricultural investment.

VL  - 9
IS  - 2
ER  -

Copy | Download

Author Information

Shafe Zelalem Gasisa

Natural Resource Economics and Management, Assosa Agricultural Technical and Vocational Education Training College, Assosa, Ethiopia

Contact Email
Mihret Fentahun Yeneneh

Land Administration and Management, Assosa Agricultural Technical and Vocational Education Training College, Assosa, Ethiopia

Contact Email

http://orcid.org/0009-0000-1233-6132
Teha Romanu Benti

Agricultural Economics, Assosa University, Assosa, Ethiopia

Contact Email

http://orcid.org/0009-0002-2949-7017

Download PDF

Figure 1

Figure 1. Propensity score spreading and common provision for propensity score estimation.

Plain Text BibTeX RIS

APA Style

Gasisa, S. Z., Yeneneh, M. F., Benti, T. R. (2024). Effects of Large-Scale Agricultural Land Investment on Local Communities Livelihoods: Evidence from Bambasi Woreda, Western Ethiopia. American Journal of Environmental and Resource Economics, 9(2), 20-39. https://doi.org/10.11648/j.ajere.20240902.11

Copy | Download

ACS Style

Gasisa, S. Z.; Yeneneh, M. F.; Benti, T. R. Effects of Large-Scale Agricultural Land Investment on Local Communities Livelihoods: Evidence from Bambasi Woreda, Western Ethiopia. Am. J. Environ. Resour. Econ. 2024, 9(2), 20-39. doi: 10.11648/j.ajere.20240902.11

Copy | Download

AMA Style

Gasisa SZ, Yeneneh MF, Benti TR. Effects of Large-Scale Agricultural Land Investment on Local Communities Livelihoods: Evidence from Bambasi Woreda, Western Ethiopia. Am J Environ Resour Econ. 2024;9(2):20-39. doi: 10.11648/j.ajere.20240902.11

Copy | Download

@article{10.11648/j.ajere.20240902.11,
  author = {Shafe Zelalem Gasisa and Mihret Fentahun Yeneneh and Teha Romanu Benti},
  title = {Effects of Large-Scale Agricultural Land Investment on Local Communities Livelihoods: Evidence from Bambasi Woreda, Western Ethiopia
},
  journal = {American Journal of Environmental and Resource Economics},
  volume = {9},
  number = {2},
  pages = {20-39},
  doi = {10.11648/j.ajere.20240902.11},
  url = {https://doi.org/10.11648/j.ajere.20240902.11},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ajere.20240902.11},
  abstract = {Most of the African countries including Ethiopia are often known by problem of large scale agricultural land investment effects on the local community where land is being transferred to investment. Even though several efforts made so far to solve the overall effects of large scale agricultural land investment situation, the challenge is still widespread problem in Ethiopia. Hence the study project's purpose was to find out how large agricultural investments in Bambasi Woreda, Western Ethiopia impact the livelihoods of the surrounding populations and examine the local communities' participation in the large scale agricultural investment. In order to attain these objectives, data were collected from 330 randomly selected households in four purposively selected kebeles of the district for both control groups and treatment groups. The sample size was chosen using a multistage stratified random sampling technique. Both qualitative and quantitative data gathering techniques and instruments were employed in the study. Besides, the instruments utilized to collect the data were observations, focus groups, interviews, household surveys, and document reviews. Data was analyzed using both descriptive statistics and econometric methods. The study shows that out of the total sampled respondents 30 of the treatment group and 13 percent of control group reveal that the project provided opportunity in terms of employment opportunity, technology transfer, utilization of agricultural inputs, changing the working culture of the community and productivity. The chie square value shows there is statistical significance among treated and control group on opportunity investment provided for the household and community. A binary logit regression model was used to describe how large-scale agriculture land investment affected the local community's standard of living. The findings indicated that only six variables were found to be significant out of the characteristics that were expected to influence local community employment in large-scale agricultural land investment projects. These includes household's educational accomplishment, size of HH, occupation of HH, Loss of useful land due to investment Project and technology transfer significantly and positively affected the employment opportunity in large scale agricultural investment projects, whereas the distance of a household's home from an investment project has a negative impact. Large-scale agricultural investments have a detrimental influence on household wealth accumulation and income, according to the estimation results of the average treatment effects on the treated. The management and implementation of land transfer for large-scale agricultural investment projects is inadequate, lack of openness, absence of community consultation, natural forest degradation, socio-economic and ecological effects must be carefully considered before transferring the land for large-scale agricultural investment.
},
 year = {2024}
}

Copy | Download

TY  - JOUR
T1  - Effects of Large-Scale Agricultural Land Investment on Local Communities Livelihoods: Evidence from Bambasi Woreda, Western Ethiopia

AU  - Shafe Zelalem Gasisa
AU  - Mihret Fentahun Yeneneh
AU  - Teha Romanu Benti
Y1  - 2024/05/10
PY  - 2024
N1  - https://doi.org/10.11648/j.ajere.20240902.11
DO  - 10.11648/j.ajere.20240902.11
T2  - American Journal of Environmental and Resource Economics
JF  - American Journal of Environmental and Resource Economics
JO  - American Journal of Environmental and Resource Economics
SP  - 20
EP  - 39
PB  - Science Publishing Group
SN  - 2578-787X
UR  - https://doi.org/10.11648/j.ajere.20240902.11
AB  - Most of the African countries including Ethiopia are often known by problem of large scale agricultural land investment effects on the local community where land is being transferred to investment. Even though several efforts made so far to solve the overall effects of large scale agricultural land investment situation, the challenge is still widespread problem in Ethiopia. Hence the study project's purpose was to find out how large agricultural investments in Bambasi Woreda, Western Ethiopia impact the livelihoods of the surrounding populations and examine the local communities' participation in the large scale agricultural investment. In order to attain these objectives, data were collected from 330 randomly selected households in four purposively selected kebeles of the district for both control groups and treatment groups. The sample size was chosen using a multistage stratified random sampling technique. Both qualitative and quantitative data gathering techniques and instruments were employed in the study. Besides, the instruments utilized to collect the data were observations, focus groups, interviews, household surveys, and document reviews. Data was analyzed using both descriptive statistics and econometric methods. The study shows that out of the total sampled respondents 30 of the treatment group and 13 percent of control group reveal that the project provided opportunity in terms of employment opportunity, technology transfer, utilization of agricultural inputs, changing the working culture of the community and productivity. The chie square value shows there is statistical significance among treated and control group on opportunity investment provided for the household and community. A binary logit regression model was used to describe how large-scale agriculture land investment affected the local community's standard of living. The findings indicated that only six variables were found to be significant out of the characteristics that were expected to influence local community employment in large-scale agricultural land investment projects. These includes household's educational accomplishment, size of HH, occupation of HH, Loss of useful land due to investment Project and technology transfer significantly and positively affected the employment opportunity in large scale agricultural investment projects, whereas the distance of a household's home from an investment project has a negative impact. Large-scale agricultural investments have a detrimental influence on household wealth accumulation and income, according to the estimation results of the average treatment effects on the treated. The management and implementation of land transfer for large-scale agricultural investment projects is inadequate, lack of openness, absence of community consultation, natural forest degradation, socio-economic and ecological effects must be carefully considered before transferring the land for large-scale agricultural investment.

VL  - 9
IS  - 2
ER  -

Copy | Download