Determinants for the escalation of informal settlements and its consequences in the suburbs of Butajira Town; Central Ethiopia

ABSTRACT Informal settlement is a global problem practised by all social classes. However, the extent and the context of the problem are serious in developing countries. The objective of this study was to identify the determinant factors for the escalation of informal settlements and their consequences in Butajira town, central Ethiopia. A mixed research approach with a sequential explanatory design was employed. Both qualitative and quantitative data were systematically collected and analysed using softwares like ArcGIS 10.3, ENVI 5.3, and SPSS. Primary data was collected through a household questionnaire survey from 221 samples and thirteen key informant interviews. Secondary data was also generated from different governmental reports, previous studies, and books. Employing a multiple linear logistic regression analysis, the empirical findings revealed that the informality was determined by a combination of factors such as monthly income, previous residence, mode accession of the respondent’s current holding, and brokers’ agitation as they were found to be statistically significant in determining the informality at a 95% confidence interval. It is found that informal settlements are expanded alarmingly in the study areas at the expense of the nearby agricultural fields. The effects are witnessed in the town resulting disorganised infrastructures and a spontaneous network of villages. The Municipality is engaged in preventive measures like demolitions and punishments but it did not bring any meaningful solutions. The researchers recommend the ‘punitive approach of marginality theory’ that insists on the forceful relocations of the informal settlers, but with the provision of adequate services


Introduction
Urban informality is defined as holding plots of land for settlement outside of the legal environments and established norms (López et al. 2019).It is a multifaceted process demonstrated by the absence of ownership rights, lack of formal planning requirements, no access to infrastructures and being environmentally unsuitable for settlement (Onyekachi 2014; UN-HABITAT 2016).Informal settlements sometimes are also sites of significant environmental risk in developing countries (Deeyah et al. 2021).
Though urban informality is a global problem (Matamanda et al. 2020), it is very serious in the global south (López et al. 2019).Today, 54% of the world's population lives in urban areas, where one-fourth of them are living in informal settlements (Konishi 2015).The speed and scale of urbanisation bring challenges.Demands for housing, basic services, functional transport systems, and jobs continue to surge, and as cities fail to keep pace with the rapid urbanisation, informal settlements grow (UN 2014;Konishi 2015;López et al. 2019).
When it comes to Africa, currently, 40% of the population lives in urban areas.This figure is expected to reach 54% by 2030, meaning that the urban population of the continent would likely triple over the next 40 years (UN 2014).A marked phenomenon of rapid urbanisation in Africa has been the proliferation and uncontrolled spread of spontaneous or informal settlements, and currently, around 90% of the residents are live in an informal settlement category (UN 2014; López et al. 2019;Matamanda et al. 2020).Adegun (2019) and Alabi (2019) found that almost half of the urban communities in the Sub-Saharan region live in the areas that are categorised as informal settlements.The rate of urbanisation in East Africa is very high (more than 4.5% on average) and at the same time it is accompanied by high rate of urban informality as well (Masselière 2018).The cause of high informal settlement in developing countries is usually associated with the high rate of rural-urban migration, the impossibility of the government to provide housing, economic hardships to afford the high urban lease prices, and the lengthy bureaucracy of gaining urban land (UN HABITAT 2016).
Ethiopia is among those countries which are characterised by a high rate of urbanisation and informal settlement expansion.Rapid urban growth is occurring in Ethiopia which resulted in pressure on jobs, services, and the like (Admit et al. 2009).The process of urbanisation in Ethiopia involves the conversion of agricultural land to urban areas (Sayeh et al. 2016).This fact can also be assured by what is being practically observed in different towns and cities of the country (Ibid).For instance, the study of Adegun (2019) and Hoeltl et al (2020) found that around 80% of the population in Addis Ababa, the capital of Ethiopia is estimated to live in informal dwellings.Besides the horizontal expansion of the major cities, many new cities are also emerging in different parts of the country (Sayeh et al. 2016).Land invasion and the rapid growth of informal settlements were observed in Ethiopian urban centres, especially in the periurban areas (Adam 2014;Tura 2018).Therefore, urban areas in Ethiopia are expanding at fast rates and this expansion is deemed to be more even in the future (Sayeh et al. 2016).With 22.2% population lives in urban areas, Ethiopia has 4.8 urban growth rate (worldbank.org).
There are multi-dimensional causes for the prevalence of informal settlements.But such causes are different across countries (López et al. 2019).Among the causes of informal settlements in Ethiopia, high rate of population growth, rural to urban migration, and the inability of the local government in providing the land and housing necessities are some of the major ones (Abagissa 2019).Akirso (2020) identified that shortage of residential housing, the increasing price for house rent, the need to have large plots of land, and the inefficient land administration system are among the major causes for the escalation of informal settlements, especially in Jimma town.Large plots of land are always demanded by most people for various reasons.The first is such plots are sources of wealth.The second is the historical and psychological attachments of the people to control large plots of land.The institutional dichotomy between the urban and rural land administration is mentioned as one of the causes which led to the expansion of informal settlements, especially in the peri-urban areas (Nega et al. 2021).In addition, Adam (2014), also reported that the inability of people to afford the increasing urban lease system by the residents and the insensitiveness of housing projects like condominiums to the needs of the people are the major reason for the proliferation of new informal settlements in the peri-urban areas of Ethiopia.
The escalation of informal settlements will have manifold legal, economic, social, environmental, and political implications.Some of these as summarised by Abagissa (2019) include health problems, environmental deterioration, social distress, crime, and urban violence.Adam (2014) and Adegun (2019) also noted that the sprawl of informal settlements will have adverse effects on the environment, and the socioeconomic well-being of urban dwellers more than the destitute image of the town under development.Hoeltl et al (2020) stated that the increasing trend of informal settlements in Ethiopian cities has resulted in poor infrastructure.And most importantly, the rapid urbanisation process in Ethiopia is taking place at the expense of farmland and forests (Admasu et al. 2020).
The existing paradigms regarding informal settlements are not to eliminate them, rather it favoured upgrading and formalisation, and that is why the tendency of informality is increasing through time (López et al. 2019).In some areas, however, the urban land laws in Ethiopia fails to deal with the formalisation of informal residential holdings.Abebe et al. (2019) identified that informal settlements in Ethiopia are increasing after the lease proclamation of 724/2004.Though informal settlements account for up to 30% of the residential holdings in Addis Ababa, the capital of the country, no policies or procedures require the systematic regularisation of informal holdings.Formalisation projects have no basis in federal legislation, and the few sporadic initiatives to formalise existing settlements in Addis Ababa, Dire Dawa, and Hawassa were very costly and of a discretionary nature.It was implemented by ad hoc municipal directives, which lacked transparency and were discontinued without reaching their targets (Deininger et al, 2012).
There is no consensus on the roles, features, and paradigms of urban informality.Some scholars contemplate informality from the point of view of prourban development while others view informality against of urban development.For example, from an economic point of view, there are two arguments given with respect to the roles of informal settlements to the residents.The first argument is related to the positive roles that are explained with the existing affordable housing, food, increased labour supply, and, low-cost gateways to economic opportunities in cities which paves better employment, especially for the migrant population.On the contrary, such places could be the centres of 'poverty traps' due to insecurities, vulnerabilities, the harsh environment, and stiff competition for survival as part of the negative argument (Turok and Borel-Saladin 2018).Turner (1963) called informal settlements 'barriadas' which have a significant contribution for the urban poor.Consolidation and regularisation programmes need to be adopted in such areas.In most developing countries like Ethiopia, Kenya, Ruanda, Nigeria etc. the practice of informal settlement is not welcomed and many of the settlements are built on the unsuitable areas.Government policies are geared towards demolishing than regularisation in most cases (Wakhungu et al, 2010).
In Addis Ababa, though recurrent reactive measures of tasks to curb informal settlements are implemented, regularisation of informal settlements is also done.There was a directive issued in 2006 to formalise the informal built-ups that are constructed in the city before April 2005 (AACA 2006).Similar attempts are also considered in other regional government's city administrations.But the directive failed to bring sustainable solutions and the trend of informal settlements becomes sporadic which makes forced evictions and demolishing common, especially in Addis.More people, including the rich, are engaging in construction of informal settlements by speculating that they would be regularised in the future.The future value of the land after regularisation is always the one that attract people in this business.
Many international organisations and governments are accepting the upgrading of informal settlements at least theoretically through implementation limitations (Matamanda et al. 2020).
Albeit a large number of studies are conducted on the causes and consequences of urban sprawl and informality (Abebe et al. 2019;Abagissa 2019;Baye et al. 2020;Akirso 2020), few studies are available on determining factors that contribute to the escalation of informal settlements, especially in the global south albeit a high rate of informality.Some of them are discussed in the following paragraph.Alabi (2019) which is focused on the determinant factors for the expansion of informal settlements in sub-Saharan African countries by taking Nigeria as a case study and found that various socio-economic, cultural, physical, and historical factors influenced the growth of informal settlements.Ezebilo and Savadogo (2021) conducted a similar study in Papua New Guinea and found that house and land ownership, frequency of crime in the area, household size, occupation, access to the toilet, and the number of years lived in the informal settlement are the drivers for informal settlements.
Due to various technical and political factors, an updated and structured study regarding the extent and context of informal settlements is lacking in Ethiopia (Negese 2020).Urban growth studies in most aspects concentrate on big cities and metropolises.Nevertheless, medium and small urban regions may possess maximum urban growth rates in a specific time interval from their establishment time (Dadras et al. 2014).Recently, many research papers have been conducted on change detection, especially on land use and land cover changes in Ethiopia.For instance (Sayeh et al. 2016), have examined the characteristics of urban expansion and land use and land cover change and its effects on the land tenure security situation of the suburb population of Debre Markos and Bahir Dar cities. Imam et al. (2013) studied the consequences of urban sprawl on agricultural land loss.Adane (2018) studied urban sprawl and informal settlements in Wolkite town, Ethiopia.The environmental impacts of urban land use and land cover changes were also studied by Bamlaku (2009).Therefore, to the best of the researcher's knowledge, though there are studies on the causes and consequences of informal settlements in some parts of the country, the determinant factors for the escalation of informal settlements are not well-researched in the Ethiopian cities' suburbs.
In cognisant of this research gap, the focus of this paper, therefore, is to identify the contributing factors for the escalation of informal settlements, analyse the consequences, assess the measures taken by the city administration to undertake the problem, and examine the spatial and temporal land use land cover change of Butajira town for the last 21 years, since 2000 to 2021.
Besides knowledge transfer, this study will be very helpful for urban developmental planners, environmentalists, policy advisors, and policymakers to design appropriate interventions in the future land use land cover pattern in the study area and to improve the urban sprawl in controlling informal settlements.The output of the study will benefit the local community with the interventions that will be done by planners, environmentalists, and or concerned government bodies.From a policy perspective, though the study is delimited in a narrow case study, it will shed light on the trend of informality, especially in the rapidly growing towns and cities.

Description of the study area
Butajira town was founded in 1936.It is one of the reform towns in the Southern Nations Nationalities and People's Region (SNNPR) and it has a town administration (municipality) which is organised into five Kebeles. 1 The town is located in the Gurage Zone, Meskan Woreda, at a distance of 135 km from Addis Ababa and 163 km from Hawassa, the regional capital and 96 km from Wolkite, the zonal capital (Gurage Zone Department of Culture, Tourism and Governmental Affairs 2012).Its geospatial location is 7º 71'to 8º 45' Latitudes & 37º 99'to 37º 71' Longitudes (Figure 1).Butajira had its own town administration with a structural plan since the reform of 1998 in Ethiopia.
The total population of Butajira according to the report of CSA (2007) was 33,406.Among these 16,923 of them were males and the rest 16,483 were females.In the same year, the numbers of households were 8,565.According to the municipaly's report, various ethnic groups are living in the town mainly the Sebat Bet Guraghes, 2 the Amharas, the Oromos and others.In terms of religious composition, the majority of the residents are Muslims followed by Protestants and Orthodox Christians.The town is organised into two Kifleketemas (sub-cities), five Kebeles, and 57 villages.

Research approach and study design
The study employed a mixed research approach with a sequential explanatory research design to answer the research objectives because the problem under investigation has both quantitative and qualitative aspects of issues.A mixed research approach enables the investigators to use both quantitative and qualitative data because they are important to provide the best understanding of a research problem and answer research questions.It will also open the door to multiple methods, different worldviews, and different assumptions, as well as to employ different forms of data collection and analysis methods.It employs strategies of inquiry that involve collecting data either simultaneously or sequentially to best understand the research problem.The data collection also involves gathering both numeric data (e.g. on questionnaires) as well as text information (e.g. on interviews) so that the final database represents both quantitative and qualitative information (Creswell 2003).The residents of the town, officers from the municipality, and elders from the study areas participated in the study.The residents who engaged in the construction of informal settlements were addressed through a self-administered questionnaire by means of an interview and the officers from the municipality and elders from the community were addressed by an interview.The reason for employing a self-administered questionnaire was that most of the residents in the suburbs were not able to read and write.

Types and sources of data collection
In addition to the data generated from GIS and the remote sensing sources, socio-economic quantitative and qualitative data were collected through using a survey questionnaire, key informants' interviews (KII), and secondary sources.

Household Questionnaire Survey
Among the most important data collection tools, which are used in the undertaking of this research household questionnaire survey is one of them.A self-administered questionnaire was conducted with the help of three assistant data enumerators.The reasons for employing a self-administered questionnaire were due to most of the residents in the suburbs not being able to read and write, and for better quality data.Hence, a total of 221 individuals were able to participate in the questionnaire.

Key Informants Interview
The key informant interview (KII) is used to assess issues that were not thought to be answered through the questionnaire as well as for triangulation and exploring further information about the issue.Officers from the municipality of the town, land brokers, and elders who know about the expansion of informal settlements in the area were involved in the research.In this regard, it was possible to contact a total of thirteen key informant interviewees in this regard.

Spatial data acquisition and processing
Multi-temporal Landsat images were used for the year 2000, 2016, and 2021.Apparently, Landsat7 ETM+ was used for the 2000, SPOT for the year 2016, and Sentinel for 2021, which is acquired from the United States Geological Survey (USGS), the Ethiopian Mapping Agency (EMA), and the European Space Agency (ESA) respectively, and projected to UTM coordinate system, map zone 37° N of clack 1880 spheroid and Adindan datum.The satellite images were geo-referenced using the topographic map.The neighbourhood resampling technique was used to have the same resolution so that accurate classification and change detection can be made.The remote sensing images were clipped using the study area.Table 1.Imagery data used for the study.
For the year 2000 satellite image, Bands 1-7 were used for calibration, but to reduce noise and disturbance in the image, band six was not included in the calibration.Then, layer stack was done to ease the classification.Unsupervised classification (ISO-DATA) was done before the supervised classification to know the number of classes able to generate from the map (Figure 2).The same procedures were applied to the satellite images of 2016 and 2021 to improve the quality of the images for better interpretation and classification.
After atmospheric, radiometric, and geometric corrections for satellite images, the maximum likelihood supervised classification method was used since it assists in the classification of overlapping signatures; pixels were assigned to the class of highest probability.Accordingly, three land use land cover (LULC) types were developed namely, open-land, built-up, and vegetation.Open-land is the type of land use that is open, and no activities are underway.It may also include, barren lands, and reserved areas for future expansion.Built-up is a land-use type referring Consulting with the local community during the mapping process in the form of Participatory Rural Appraisal (PRA) was one of the solutions taken by the researchers to ease the classification.A total of 150 ground control points (50 per class) were collected through GPS for calculating the accuracy assessment for each year's satellite image.Accuracy assessment then was calculated by crosschecking the collected GPS points with the classified images and it was found 80% for 2000, 82% for 2016, and 88% for the 2021 images.

Sample size determination and sampling methods
The study population consists of those individuals who are involved in the construction of informal settlements, where the list was taken from the municipality.Purposive and simple random sampling techniques were employed to select the target Kebeles and the questionnaire respondents respectively.Firstly, three Kebeles out of the five Kebeles were selected purposively.Kebeles with relatively large numbers of informal settlements and the availability of data and address of informal settlers are some of the reasons for taking the samples.Then, questionnaire respondents were selected through simple random sampling.The sample was drawn from Kebele 03, 04, and 05 because these Kebeles have a relatively higher number of informal settlers than others Table 2.The finite population formula of Israel (1992) was employed to determine the sample size.The total number of informal settlers in these three Kebeles is 593.
For a confidence limit of 95%, a probability error of 5%, and an estimated proportion of the population, the sample size of the study was calculated to be 221.The sample respondents were selected randomly out of the 593 informal settlers through random sampling.
Accordingly, as per the number of informal settlers in respective Kebeles, proportional samples were drawn from the selected three Kebeles.Therefore, 79, 69, and 73 sample sizes were taken from Kebele 03, 04, and 05 respectively.

Methods of data analysis
Both quantitative and qualitative data analysis techniques were employed.The quantitative data were analysed through a binary logistic multiple linear regression model through Statistical Package for Social Sciences (SPSS) version 21.The technical GISbased and remote sensing data was analysed through ArcGIS 10.3 and ENVI 5.3 software.The qualitative and quantitative data were analysed sequentially by the logic of sequential explanatory design.This quantitative dominant research design gives freedom to collect and analyse the quantitative data first and then followed by the qualitative one (Creswell 2009).This research was performed in line with the following underlying flowchart (Figure 2).

Variables and the model
Binary logistic multiple linear regression models were used to present the results of the data and to show the relationships between and among different dependent and independent variables.This model was selected because the dependent variable (informal settlement) is a discontinuous dichotomous variable and there are both continuous and discontinuous independent variables that are supposed to influence the dependent variable.The household's nature of settlement (informal settlement) is a dependent variable, and the following are some of the independent/explanatory variables.Annexes (1) X 1 : Age of the respondent (2) X 2 : Educational status of respondent (0: Not read and write, 1: Read and write) (3) X 3 : Household family size (4) X 4 : Occupation of a respondent (5) X 5 : Respondents' monthly income (6) X 6 : Residential landholding size of respondents in m 2 (7) X 7 : Previous residence of the respondent (0: Outside Butajira, 1: Butajira) (8) X 8 : Respondents mode accession of their current holding (0: Buying from farmers, 1: Lease holding, 2: Gift, 3: Non-lease municipal holding) (9) X 9 : Broker's agitation on respondents to buy informal land (0: No, 1: Yes) (10) X 10 : Responsiveness of the municipality towards formal land access requests from respondents (0: No, 1: Yes) These variables are developed from the literature and the responses of the respondents during the pilot study.Peer review was done to check the appropriateness and maintain the standard of the variables before the data collection.
The above selected 10 explanatory variables can be put in a regression formula like Y ¼ βX1 þ βX2 . . .::βX10 (when all are significant) (2) Where Y is the dependent variable β Is the constant and X is the explanatory variables

Results
Before proceeding to the detailed data analysis and interpretation, it is better to present the descriptive statistics of the respondents.Table 3 summarises the background profiles of the respondents.Accordingly, among a total of 221 respondents, about 61.8% were male and the remaining 38.2% were female informal settlers.The average age was found to be 43.09years.When it comes to their marital status, 77.8% were married and only 22.2% were single.Only 19% of them were able to read and write whereas 71% of them were not able to read and write.The average number of children was 2 which is lowest than the national average, 4.014 (UN 2019).When it comes to their occupation, while only 16.3% were civil servants, the remaining 83.7% are farmers, merchants, and other unemployed.The average daily income was 1.6 $, which is below the international poverty line of 1.9 $ (UNCTAD 2021).This means that the informal settlers are in absolute poverty and it will have some push factor for their involvement in the construction of houses informally.An option was also given for the respondents to know about their previous residence, attesting that 76% were migrants from other rural areas and towns whereas only 24% of the informal settlers respond that they previously lived in Butajira town.The majority of respondents (86.9%) answered that they accessed their holdings by buying from farmers.For the last 30 years, vegetation cover is depleted continuously in Butajira.It was 1001.3 ha (37.5%) in 2000 and reduced to 628.2 ha (23.5%) in 2016 and only 160.4 ha (5.9%) in 2021.However, open land is increasing from 1191.9 ha in 2000 to 1768.4 ha in 2016 and 1807.2 ha in 2021.Built-ups show a decrease in the first analysis period (2000 to 2016) from 17.8% to 10.18%, but it increases from 10.18% to 26.4% in the second analysis period (2016 to 2021) (Table 4).

LULC analysis
As indicated in Table 5, the area covered by vegetation and built-ups encountered −1.2% and −1.4% annual conversion, respectively, for the first analysis period (2000 to 2016).Open land increased by 1.6% per annum in the same year.It is surprising to observe the conversion of 271.5 ha of built-up area in 2016 to 714.8 ha in 2021 (Table 4) which is (about a 443.3%rate of change) with an average annual rate of 29.3% per year.Later on, in the second and third analysis periods, built-ups and open land show a continuous increase at the expense of vegetation.

Determinants of informal settlement: multiple binary linear logistic regression analysis
Among the various causes, which affect the escalation of informal settlements, the determinants i.e. the most important of them need to be identified by regression analysis.To do so ten independent variables, which have a possible influence on urban informality, were adapted from the literature.
As presented in Table 6, multiple linear logistic regression analysis for Butajira town, among the ten independent variables, four of them, i.e., monthly income, previous residence, mode of the accession of the respondent's current holding, and brokers agitation were found to be statistically significant in determining the respondent's informal settlement at 95% level of significance.
According to the results of the regression analysis, among the significant explanatory variables, the previous residence of the respondents was 24.1% times more likely to predict the informal settlement.Monthly income predicts informal settlements with 0.999% and brokers' agitation predicts informal settlements with 0.146% times.
The final regression equation is written as; The previous residence of the respondents was found to be one of the determinant factors for the escalation of informal settlements in Butajira Town.Those respondents who were living inside and those who migrated from other areas of the town are all engaged in the construction of informal settlements.And it is noted in the profile of the respondents that the previous residents of the majority of the informal settlers (76%) were migrants from other rural areas or other nearby towns.In another response, 86.9% of the respondents said that they accessed their current holdings through purchasing from farmers (Table 1) mainly with the broker's agitation.It shows the brokers had a considerable role in influencing the households to buy informal houses.

Causes and consequences of escalation of informal settlements in Butajira
In addition to the regression model test to identify the determinant factor, an attempt was also made to identify the possible causes of increasing informal settlements in the study areas.
The household questionnaire survey revealed that informal settlements in the study area are the result of an overstated land lease system, rural-to-urban migration, unaffordable land price and high house rent prices.The price provided by the bidders to the lease system is very costly during the auction.These are some of the major causes for the proliferation of informal settlements in Butajira.The exaggerated prices of land lease system are not pro-poor and people prefer to engage in the informal market to afford for their housing since it is cheaper than the formal one.There is always a strong competition when urban land is open to the auction.This 'the fittest will survive' competition is difficult to large number of people and they go for cheaper options to finance their residence.Figure 4 summarises the causes behind informal settlements and the response.The data from the Key Informants Interview (KII) show additional causes for the increasing trend of informal settlements.Brokers' agitation, the involvement of richer people in the process, the low response of the municipality, and the ownership clash between the Mareko and the Guraghe ethnic groups over the territory are some of the causes behind the increasing trend of informality in the town.
Informants from the municipality responded that brokers are misleading the farmers and inspiring them to sell their plots otherwise the government will expropriate their property without compensation.The low awareness of farmers about the urban land lease laws makes them easily deceived by the brokers.Therefore, they engage in selling their plots.The question of ownership of the town is one of the causes to the expansion of informal settlements in the study area, which was mentioned by the informants during the interview.The Mareko ethnic group claims that the town and expansionary areas are belonged to them since these groups are living in the suburb areas; while the other ethnic groups like the Guraghe and other minorities are considered as immigrants .
On the other hand, some ethnically Mareko informants had different responses.They said that they are not allowed to construct buildings by the town municipality, which was dominated by Guraghe and non-Mareko groups.An interview with the head of the municipality shows that ethnic clashes over ownership rights are also common in other towns of the zone, like in Wolkite and Gubrie towns.
The consequences of informal settlements were also the objective of this study and there was a question for both questionnaire respondents and the informants.The responses of the household survey are presented in Figure 5. Accordingly, land invasion, conflicts, disorganised infrastructures, corruption and maladministration and loss of government  revenue are some of the consequences according to the household survey.

Measures taken by the city administration to curb the informality
This section illustrates the preventive and remedial actions that were taken by the local government to reduce the problem of informality.Table 7 summarises the questions for the types of interventions made by the municipality to curb the informality.The Municipality did not almost practice preventive measures (issuing directives, awareness creation, and strict policy measures) which are taken to prevent the emergence of illegal settlements.However, corrective measures seem practical in the study area as reported by the majority of the respondents, which contains 68.8% demolitions, 15.8% upgrade and formalisation, and 12.7% punishments.
The measures taken by the concerned government organ were also a subject of discussion with the key informants.Almost all of the informants agreed that there were no directives and regulations issued concerning informal settlements and people were engaged in this activity by hoping that it will be upgraded and formalised at the end of the date.One of the informants said the following; 'I have invested everything in this house.But now I left with nothing.They ruined everything.It was good if they told me that it is illegal to construct houses in this area.Nobody told me that my holding is an illegal one' (KII from informal settlers, 2021).It shows how curative measures like demolitions are very costly and discretionary.

Discussion
As indicated by this study and other similar studies in Ethiopia, informal settlements are expanding at fast rates with the expense of the nearby agricultural lands and forests (Adam 2014;Sayeh et al. 2016;Tura 2018;Admasu et al. 2020;Dires et al. 2021).
The main reason for this proliferation is the population pressure-induced rural-urban migration, the unaffordable lease system, the inefficient land administration system, and the desire to own large plots (Adam 2014;UN-HABITAT 2016;Tura 2018;Abebe et al. 2019).The continental and regional trend of informality shows that it is increasing from time to time as a result of population pressure, the inefficiency of the Municipalities, and corruption in Nigeria, Kenya, Ruanda, Tanzania, and Ethiopia (Wakhungu et al, 2010;Ali and Sulaiman 2006;Alabi 2019;Abagissa 2019;Ezebilo and Savadogo 2021;Deeyah et al. 2021).
When it comes to the study area, the problem of informal settlement is persistent; it is mainly caused by the inflated land lease system, rural-to-urban migration, unaffordable land prices, and high house rent prices.The binary logistic regression analysis has also resulted in four significant independent variables, which determine the prevalence of informal settlements.These are monthly income, the previous residence of the respondents, mode of accession of the respondent's current holding, and brokers' agitation.
Accordingly, low-income individuals are engaged in the construction of informal settlements.The reason for this according to the data gained from the interview was the costly and bureaucratic as well as unaffordable procedures of the formal urban land lease laws.The previous residence of the respondents is also one of the significant factors determining informal settlements, in that individuals who migrated from other areas are more involved in owning informal settlements than the populaces.Respondents' mode of accession of their current holding (0: Buying from farmers, 1: Gift, 2: Non-lease municipal holding) was also one of the predictor variables and found that all of the respondents who accessed their holding in the three listed modalities are engaged themselves in the informal settlement.Brokers are playing a critical role in misleading respondents to buy informal houses.
Previous studies are also congruent with this study concerning the causes of informal settlements.For instance, Abagissa (2019) found that rural-urban migration is one of the causes behind the proliferation of informal settlements in Addis Ababa.Adam (2014) also mentioned the non-affordability of the current lease system as one of the causes.Hoeltl et al. (2020) also reported similar findings in that the impossibility of the government to provide housing, economic hardships to afford the high urban lease prices, and the tiresome bureaucracy of gaining urban lands.The study of López et al. (2019) reveals a different result in that 'while urban informality is usually associated with widespread poverty, current trends show that this is not always the case' (p.9).Similarly, Nega et al. (2021) found that the proliferation of informal settlements and land invasion in the peri-urban areas of most cities and towns of Ethiopia are contributed by the institutional dichotomy of the rural and urban land administration sectors.The increasing trend of informal settlement in the study area is associated with the high rate of rural-urban migration, the impossibility of the government to provide housing, economic hardships to afford the high urban lease prices, and the onerous bureaucracy of gaining urban lands.This finding is similar to the study of Hoeltl et al. (2020) and Adam (2014).Deininger et al. (2012) conducted a study on the status of informal settlements in three Ethiopian cities, Addis Ababa, Dire Dawa, and Hawassa, and quantified the percentage of informal settlements in which 30% of the residential holdings; especially in Addis are categorised as informal.Adane (2018) has studied urban sprawl and informal settlements in Wolkite town, Ethiopia, and found that the monthly income of respondents, previous residence, occupation, and mode of accession of respondent's current holding is found to be the factors determining the nature of the settlement.This shows how the problem of informality is serious in both the federal and the regional cities. Abebe et al. (2019) showed the increasing trend of informal settlements in Jimma town using GIS and remote sensing.They found that informal settlements are increasing at the expense of the nearby agricultural lands in Jimma town, which is similar to Butajira.
Concerning the consequences of informal settlements, there are related studies that are corresponding with the findings of this study.In this regard, Onyekachi (2014) explained the consequences of informal settlements in a vicious circle.Accordingly, poor income leads to poor health, which is attributed to poor education and poor housing, then these further results in poor employment opportunities, and such problems are revolving around a circle.One of the prominent effects of the expansion of informal settlements on the LULC trend of the study area as stipulated by the remote sensing and GIS analysis shows that informality is expanded with the cost of the nearby agricultural fields which is supported by many researchers like Admasu et al. (2020), Adam (2014), Tura (2018) and Sayeh et al. (2016).
Informal settlements resulted in the deterioration of the image of the town.Networks of electric wires are scattered here and there, unfinished roofs and walls, and residential buildings with no fence make the village look like a 'refugee camp'.The networks of electric wires are sources of danger and risk.Generally, as far as the analysis and investigation of researchers are concerned, among the Durst and Wegmann (2017) five common typologies of urban informality practices; informality derived from code violations, informality resulting from illegal subdivision, land-Use, and zoningrelated informality, informality associated to transfer of property and informality as an infringement of property rights, the case in the study area is associated with the later one, i.e. informality as an infringement of property rights.Because most of the informal built-ups are individual transgresses from the public land and some other's property illegally.
Concerning the measures taken to curb the proliferation of informal settlements, curative measures were taken by the municipality, which includes demolitions, formalisations, and punishments; where the majority of the respondents said that demolitions have taken place.López et al. (2019) however identified somewhat different findings, which stipulate formalisation and regularisation actions are taken place by the government though it leads to other proliferation.Deininger et al. (2012) have also stated that formalisation projects were very costly and remained ineffective in Addis Ababa, Dire Dawa, and Hawassa cities of Ethiopia.
Preventive measures, which are cost-effective and sound, were neglected in the study area.
Though policies or procedures that require the systematic regularisation of informal holdings are out of the scope of this study which needs further research, few sporadic initiatives to formalise existing settlements were very costly and of a discretionary nature.The possible impacts of increasing informal settlements on access to infrastructure, health, and future LULC changes in the study area also need further investigation.

Conclusion and implications
The study aimed to systematically identify the determinant factors for the escalation of informal settlements in the suburbs of Butajira town.In cognisant of this, the authors have assessed the consequences of informal settlements and examined the spatial and temporal land use and land cover changes that happened in the study area from 2000 to 2021.
The satellite imageries with GIS and RS analysis tools supported by an interview and a household questionnaire survey were used to study the determinant factors for the escalation of informal settlements in Butajira town for the last 21 years.The study indicated that there was LULC change conversion and modification witnessed by the results of the satellite image classification and analysis.Apparently, builtups and open-land areas show a continuous increase in the cost of the vegetation cover, especially during the analysis period of 2016 to 2021 There was a combination of factors attributed to the escalation of informal settlements in the study areas.
The empirical findings revealed that informality is influenced by a combination of factors.Among the selected 10 independent variables, four of them are found to be statistically significant at a 95% confidence interval.These are the respondents' monthly income, the previous residence of the respondent, their mode of accession of current holding, and the broker's agitation on the respondents to buy the informal land.
The interview data conducted with the community elders and officers from the Municipality shows similar results.Most of the residents hold land through informal transactions with nearby farmers through the help of land brokers.Among the samples taken, 86.9% of respondents get their current holding through informal transactions with the deception of land brokers.The land lease system, rural-to-urban migration, unaffordable land prices, and high house rent prices were some of the causes of informal settlements as per the household survey.
The researchers observed that there were plenty of residential buildings and edifices here and there in the Northern and Western parts of the town.Land invasion, conflicts, disorganised infrastructures, corruption and maladministration, and loss of government revenue are some of the consequences.The absence of plainly demarcated boundaries and a master plan is a challenge to controlling the spontaneous flow of settlements.The municipality was demolishing informally established houses since last June 2015.But it led to a conflict between the community and security forces and the demolishing process was abandoned with the loss of two individuals, one civil and one a member of the police force.
Based on the results of the study in general, the researchers would like to propose the following possible suggestions to control the expansion of informal settlements and for the sustainable development of the town.
• The municipality of the town should establish a clear set of guidelines and set up criteria to monitor and regulate the sustainable development of the town.• It is better to adopt preventive measures before reacting because it has socio-economic costs on society and the nation at large.More specifically, the researchers recommend the punitive approach of marginality theory of (López, Nieves & Bartolomei, 2019) to be adopted in Butajira town to curtail the future tendency of informality through forceful relocations of informal settlers but with the provision of basic services.

Multicollinearity test (needs numbering)
Linear regression assumes that there is little or no multicollinearity in the data.Multicollinearity occurs when the independent variables are not independent from each other.A second important independence assumption is that the error of the mean has to be independent from the independent variables.Multicollinearity might be tested with 4 central criteria: The first thing is correlation matrix -when computing the matrix of Pearson's Bivariate Correlation among all independent variables the correlation coefficients need to be smaller than 1.The second one is tolerance which measures the influence of one independent variable on all other independent variables; the tolerance is calculated with an initial linear regression analysis.Tolerance is defined as T = 1 -R 2 for these first step regression analysis.With T < 0.1 there might be multicollinearity in the data and with T < 0.01 there certainly is.
The third criteria is Variance Inflation Factor (VIF which is defined as VIF = 1/T.Similarly with VIF > 10 there is an indication for multicollinearity to be present; with VIF > 10 there is certainly multicollinearity in the sample.
The fourth criteria are Condition Index; it is calculated using a factor analysis on the independent variables.

Test of homoscedasticity
Test of homoscedasticity describes a situation in which the error term (that is, the "noise" or random disturbance in the relationship between the independent variables and the dependent variable) is the same across all values of the independent variables.In other words it is the condition when error variance should be constant and the variance of the residuals is homogeneous across the levels of the predicted values.Hence, there was no problem of heteroscedasticity in this data and the variances of explanatory variables were found to be constant.

Figure 3
Figure3shows how Butajira Town was expanding during the study period(2000 to 2021).Accordingly, the town is lingering with the cost of vegetation cover and open lands.Urban centres are extended more than ever, especially from the period (2000 to 2016) as the town became a self-administered city in 2006.For the last 30 years, vegetation cover is depleted continuously in Butajira.It was 1001.3 ha (37.5%) in 2000 and reduced to 628.2 ha (23.5%) in 2016 and only 160.4 ha (5.9%) in 2021.However, open land is increasing from 1191.9 ha in 2000 to 1768.4 ha in 2016 and 1807.2 ha in 2021.Built-ups show a decrease in the first analysis period (2000 to 2016) from 17.8% to 10.18%, but it increases from 10.18% to 26.4% in the second analysis period (2016 to 2021) (Table4).

Table 1 .
Imagery data used for the study.

Table 2 .
The registered number of informal settlers.

Table 3 .
Profile of the respondents.

Table 5 .
Summary of LULC average change detection.

Table 6 .
Summary of logistic regression result for Butajira Town.

Table 4 .
Area coverage of LULC types and rate of change ofButajira in 2000Butajira in  , 2016Butajira in   and 2021.   .

Table 7 .
Measures taken by the city administration.
Mr. Worku Nega graduated from Bahir Dar University in 2015 with a Bachelor of Science degree in Land Administration and Surveying.He graduated with a master's degree in remote sensing and geoinformatics from Addis Ababa University in 2018.Mr.Worku Nega has been working as a lecturer by Remote Sensing and Geoinformatics in Institute of Land Administration, Debre Markos University.Mr.Worku has an interest in the research area of remote sensing and GIS application on vegetation cover, environment and climate.In addition, he is also very interested in the field of land administration.Mr.Tilahun Direshad his LL.B degree from Debre Markos University School of Law in the year 2016 and his Msc.Degree in Land Administration and Management specialization in Real Property Law from Bahir Dar University Institute of Land Administration in the year 2018.He has been working as Lecturer of Real Property Laws and Land Management in Debre Markos University Institute of Land Administration.Besides, he is currently the Executive Director of Administrative and Students' Affairs Directorate of the University.Mr. Tilahun Dires has a research interest in the areas of Land Law and Land Management.

of the model: ANOVA
Values of 10-30 indicate a mediocre multicollinearity in the linear regression variables; values > 30 indicate strong multicollinearity.Dependent Variable: nature of settlement.b.Predictors: (Constant), Age, Sex, Response of Municipality, Brokers' agitation, Educational status of respondent, Monthly income of respondent, Residential landholding size of the respondent, Mode of accession, previous residence, Household size of the respondentAs indicated in the above table, the overall regression model for variation of nature of settlement is found to be statistically significant with F ratio = 27.278 and α = 0.000.This means that there is at least one explanatory variable is different from zero and determine the change in dependent variable.