The Impact of Resettlement in Urban Market Redevelopment on Income Inequality, Its Determinants, and Implications for the Resettled Population: Applying the Kejetia New Market Exemplar, Ghana

Abstract

Resettlement has become inevitable and necessary for urban market redevelopment to be implemented in developing countries, yet it impacts the livelihoods of the displaced population. In Ghana, studies have contributed significantly to the social processes, however, income inequality is rare. Therefore, the present study contributes to the knowledge gap by examining the impact of resettlement on income inequality in Kejetia market redevelopment in Ghana, and discusses their determinants and the implications for traders and drivers before, during, and after resettlement. Mixed methods were adopted using semi-structured interviews for field data. Theil’s T-Statistic was explored to measure inequality into three decompositions; ‘between-groups,’ ‘within-groups,’ and ‘overall’ inequalities to understand its form and trends across the three resettlement stages. The study findings showed complex inequalities in forms and trends at each stage. In addition, the findings lay bare the determinants and implications of income inequalities. Generally, the study concludes that the resettlement project not only induced unsatisfactory incomes and inequality, but equity concerns for traders’ and drivers’ groups, particularly the latter. The key implication of the research is the importance of influencing scholars and policy practitioners to give attention to the income inequality effects from resettlement in urban market redevelopment.

1. Introduction

Urban redevelopment has become an increasingly neoliberal policy driven by local governments to re-invent urban spaces and alleviate poverty [1,2]. This is a land redevelopment process that often involves the demolition of old building structures and the reconstruction of them [1,3].

Due to increasing urbanization and its subsequent challenges such as poverty, inequality, pressure on infrastructure, unemployment, etc. [4], city transformations have been fast forwarded [5,6]. The United Nations claims that more than half of the world’s population (4.2 billion) currently lives in urban areas, and projected the number to increase to 6 billion by 2041 [4]. With that said, urbanization becomes a necessary process to promote economic growth if it has the required social support and infrastructure [4,6].

Resettlement has become inevitable and necessary for urban redevelopment to be implemented [7]. The concept of Urban Redevelopment-Induced Resettlement (URIR) has both positive and negative impacts on the livelihoods of the affected populations [3,8]. That is, as the URIR provides a better standard of living and infrastructure for residents [8], adverse livelihood impacts for the resettled population are undeniable [3,8,9].

On top of that, urban marketplaces have become the key targets for URIR projects by local governments due to their central location in the commercial hub of most African cities [10]. These are open spaces with structures serving as stores and stalls for the trading of goods and services primarily for income. Therefore, as plans are made to redevelop more urban marketplaces, their livelihood impacts become a concern.

In Ghana, like other developing countries, urban market redevelopment has been contentious due to the high reliance of the urban poor, women, and the informal economy for income [11]. Local governments consider the redevelopment necessary due to the following reasons; income generation, control of fire outbreaks, flooding, and human and vehicle traffic around the marketplace [12]. However, traders often raise equity concerns, especially about the relocation sites being far-flung and non-market areas [13]. With that, the traders express their displeasure about the resettlement process through street protests and demonstrations [14,15]. Others also criticized the redevelopment as politically induced, where redeveloped stores end up in the hands of party faithfuls who in turn sell or rent it to them at high prices [13]. The Kejetia market redevelopment is a recent case example of urban redevelopment-induced resettlement in Ghana which temporarily resettled traders and drivers to other locations and returns them after redevelopment.

Previous studies have significantly contributed to the urban market redevelopment and resettlement in Ghana, mainly focusing on the sociopolitical processes or approach [10,12,13]. Asante and Helbrecht [10] analyzed displacement processes during the regeneration of the Kotokuraba market in Ghana, using Politically-induced Displacement (PID) as a theoretical construct. They concluded that PID was very pervasive in the regeneration process. From a case study of the redevelopment of the Kumasi Central Market in Ghana, Adinyira et al. [12] examined the stakeholder participation in the entire regeneration process and concluded that such participation should be recognized to ensure resilient urban regeneration. Okoye [14] also traced the fakery and Fabrication in Kumasi’s Modern Market and concluded that the term ‘Modern Market’ is fabricated by local authorities.

However, given the novelty of urban market redevelopment and resettlement in Ghana and Sub-Saharan African countries, empirical research on the impact of resettlement on income inequality is rare. The term income inequality is when a few individuals have a greater share of the total income to the disadvantage of the majority of individuals [16]. Research on income inequality is significant because when it becomes persistent, the inequality gap may widen as well as breed equity concerns for the market population. This point is reiterated by the United States Senate Bernie Sanders, saying;

“A nation will not survive morally or economically when so few have so much, while so many have so little”.

[17]

Therefore, we argue that to optimize the sustainability of urban marketplaces, there is a need for an empirical and in-depth investigation into their income inequality issues.

The present study contributes to the knowledge gap by examining the impact of resettlement in the Kejetia market redevelopment on the income inequality of the resettled population before, during, and after resettlement, and discusses its determinants and implications. Theil’s T Statistical model is used to measure income inequality across the three resettlement stages to understand the form and trend of its inequalities. The study preferred using this model due to its three-dimensional feature; ‘between groups,’ ‘within groups,’ and ‘overall inequality,’ reporting detailed and rich information on inequalities, particularly when the target population is sub-grouped [18]. Generally, the study contributes to the literature on urban redevelopment-induced resettlement, particularly urban marketplaces.

Following this introduction section, the literature review introduces the concepts and comprehensively summarized previous research on the topic, the materials and methods section elaborates on the methodology used and data collected; the results and discussion section analyzes the form and trend of income inequality across three periods in the resettlement process and discusses its determinants and implications as well, and the conclusion section summarizes and draws meaningful conclusions on the findings.

2. Literature Review

2.1. The Concept of Urban Redevelopment-Induced Resettlement (URIR)

Urban redevelopment helps to maintain a balance among social, environmental, economic, and cultural aspects to further ensure the sustainability of the redevelopment zones [1,3]. The policy is a land redevelopment process that often involves the demolition of old building structures and reconstruction of them [1].

Resettlement has become an inevitable and necessary policy for urban redevelopment to be implemented, yet causes displacement to the affected population [7]. Resettlement is defined as the process of planning for and implementing the relocation of people from one place to another, ensuring their safety and livelihood integration in the new location [19]. However, the present study extends this definition to include the return of the displaced population to their original place after redevelopment. The concept of Urban Redevelopment-Induced Resettlement (URIR) refers to the resettlement due to urban redevelopment of projects like residential and industrial buildings, infrastructures such as old railroads, open spaces like traditional marketplaces, etc. Time has shown that traditional marketplaces have been the target of URIR due to their central location in the commercial hub of most African cities [10].

2.2. Livelihood Effects of Urban Redevelopment-Induced Resettlement (URIR)

Studies show that Urban Redevelopment-Induced Resettlement (URIR) in general impacts the livelihoods of the resettled population [8,9]. Livelihood refers to the capabilities, assets (both material and social), and activities required for a means of living [3]. Gezahegn and Jan [8] claimed that URIR in a household in Addis Ababa, Ethiopia caused income loss, changing schools for children, health problems, and loss of savings. Ambaye and Abeliene [3] examined the urban relocation impacts on the livelihoods of poor households displaced by a development project in Bahir Dar city in northwestern Ethiopia and concluded that it caused a variety of livelihood risks such as loss of home or shelter, lack of urban infrastructures such as road facilities, electricity, job opportunities, potable water services and cash compensation for rehousing, as well as landlessness. According to Liu et al. [20], property-led urban redevelopment exhibits strong effects of price-shadowing and subsequent indirect displacement of the residents in the vicinity. Hubbard [21] claimed urban redevelopment induced displacement to one’s home in six London council estates affected the individuals’ psychological and physical health.

Equivalently with other livelihoods, inequality has emerged as a solid impact from URIR on the resettled population. The term inequality can be defined as the quality of being uneven and could sound similar to inequity but, there is a difference [22]. Inequity, on the other hand, refers to injustice or unfairness of some kind, when people do not get the same opportunity [22]. In that case, inequity may cause inequality concerns [22].

Egolum and Emoh [23] claimed that URIR in a developing economy induced inequality concerns such that; more homes were destroyed than were built; destroyed homes were predominantly low-rent ones; constructed buildings were predominantly high-rent ones; housing conditions were made worse for those whose housing conditions were least good, whiles improved for those whose housing conditions were best. According to Bachir and Wei [24] in their study on the socioeconomic impacts of urban redevelopment in the inner city of Ningbo, China; findings show that not all households benefit equally from residential relocation. Y. Choi et al. [25] studied the socioeconomic characteristics of residential redevelopment projects in South Korea and concluded that due to the increase in housing value after redevelopment projects, original residents are not willing to pay the high rent and rather prefer to voluntarily vacate the place whereas those with good economic status were willing to pay and stay there. Moreso, Cai, et al. [26] examined the change and disparity in the urban housing wealth of residents under redevelopment and concluded that there were downward and upward trends in housing wealth.

Previous studies have been generally on the pronouncement of the livelihood effects from the URIR, without elaborating on the underlying determinants and their subsequent implications for the resettled population. Also, the livelihood impacts of URIR on the affected population are well established but limited to income inequality, which is a key livelihood indicator as well. Most studies are more focused on residential redevelopment than the urban commercial places that have recently become a displacement hub for urban redevelopment. The present study responds to the knowledge gaps by examining the resettlement impacts on income inequality in an urban marketplace.

In Ghana, the concept of URIR takes the same rationale behind its implementation; that is, renovating or replacing dilapidated or old buildings with new housing, public buildings, parks, roadways, etc. However, in recent times, open marketplaces have become the prime target for urban redevelopment by local governments [10]. Evidence can be seen from the past Ghana Urban Management Pilot Program (GUMPP) in Kumasi, which was to identify and test systems that can be successfully replicated in other local government projects. The Kumasi city prioritized the redevelopment of three open marketplaces; Atonsu market, Old Tafo market, and Asawasi market. These projects displaced and resettled the market traders. Aside from this, several open urban marketplaces in the said region have undergone redevelopment and resettlement processes of which the Kejetia New Market is a recent case example.

Though studies are limited in the Ghanaian context, the few ones have been examining the social process of URIR, especially in marketplaces as a way of advocating livelihoods [12,14]. Adinyira et al. [12] emphasized the need for stakeholder participation in the entire regeneration process of Sub-Saharan Africa’s open marketplaces to ensure the resilience of urban regeneration. Okoye [14] also examined the fake and fabrication in urban market redevelopment, reflecting on the dispossession and displacement of the traders. She concluded that the term modern market that local authorities have been championing as if it is an inevitable urban future is fake and fabricated. Asante and Helbrecht [10] examined the processes of market redevelopment in Ghanaian cities and the politics that characterize each phase. He demonstrated that the process reflects a highly politicized nature in Ghana and Africa more broadly.

Studies in the region are not only limited to income inequality, but also pre-relocation impacts of urban market redevelopment. The present study contributes to the knowledge gap by examining the impact of resettlement in urban market redevelopment on income inequality during and after resettlement

3. Materials and Methods

3.1. Study Area

Our paper focused on the Kejetia Market as the study area, which is in the heart of Kumasi Metropolis (Figure 1), an administrative city for the Ashanti region in Ghana. The Kumasi metropolis is popularly known for its trading activities, having several satellite markets, and experiencing many cases of redevelopment-induced resettlement. However, Figure 2 shows the Kejetia market redevelopment which is a recent case example and the largest single market in Ghana with more than 8400 stores and stalls (Ambrose Agyemang, field engineer, interview 2019). For the local government to enhance the capacity to control human congestion, fire outbreaks, and flooding, the redevelopment plan was deemed necessary. The redevelopment is in three phases but the first phase is completed, and upon which the study is focused. The first phase displaced both traders and drivers temporarily in 2016, through 2019 at the relocation sites and return them after the redevelopment in 2020. The various relocation sites are additional study areas surrounding the Kejetia Redevelopment Market and they include the Racecourse, Batama market, Adehy3 market, and Abinkyi (Afia Kobi) (Figure 1). However, such resettlement projects have caused concerns about income and inequality issues.

3.2. Methodology

A case study methodology was the appropriate technique for investigation because it enables the authors to conduct an in-depth exploration of real-time phenomena within the naturally occurring context [27]. It is also suitable for research seeking to answer ‘how’ and ‘what’ questions [28], particularly the study’s questions below;

• How has income inequality changed across the three resettlement stages for traders and drivers?
• What factors account for the changes in inequality during and after resettlement?
• What are the implications for the resettled population?

Generally, our methodology consists of two main steps; in step 1, we organized a field survey of the study area (Figure 1), and in step 2, we measured and analyzed the income inequality issues for the two subgroups, using Theil’s T-Statistics models in excel spreadsheet. In the same step, we discussed its determinants and implications for the resettled population as a second part.

3.3. Step 1: Data Collection

The field survey was conducted using semi-structured interviews, particularly face-to-face and phone interviews (see Supplementary Material). Due to the in-depth information needed, face-to-face interviews were considered more appropriate [29]. Phone interviews were deemed necessary because we wanted to get the same earlier respondents to respond after the resettlement and return stage.

The study population was 2300 traders and drivers who were resettled due to the market redevelopment (Ambrose Agyemang, field engineer, interview 2019). The present study focused on the two subgroups because they are the two distinct groups in a typical urban traditional marketplace in most developing countries. These subgroups responded to questions about their monthly incomes across the three stages in the resettlement process; before, during, and after resettlement. Additionally, their perspectives on factors causing their income changes and the impacts on them were solicited as well.

Table 1. Data collection at various locations, Source: field survey 2019–2022.

Locations	Adehye3 Market	Bantama Market	Racecourse	Abinkyi	Total	Redeveloped Marketplace
Survey period	August 2019	August 2019	August 2019	August 2019		January 2022
Resettlement stage	* During resettlement	* During resettlement	* During resettlement	* During resettlement		After resettlement
Type of respondent	Traders	Drivers	Drivers	Traders		Drivers & traders
No of people	69	70	90	11	240	Same sample size
Method	Face-to-face interviews	Face-to-face interviews	Face-to-face interviews	Face-to-face interviews		Phone interviews

* Data for the ‘before’ and ‘during’ resettlement was collected at the same time.

demonstrates the data for ‘before’ and ‘during’ resettlement which was collected in August 2019 at the relocation sites, ‘whereas data ‘after resettlement and returned’ was collected in January 2022. However, their monthly incomes for each resettlement stage were collected by recording their daily incomes per week for one month.

A random sampling technique was used to sample 240 respondents from the study population due to its simplicity and lack of bias [30]. A sample size which is around 10% of the study population is considered a good maximum for a study [31]. However, the study used this sample size because it guarantees statistical significance and thus demands reporting the practical significance [32]. That is, the larger the sample size, the smaller the margin of error, controlling the risk of reporting false-negative or false-positive findings [32]. Slovin’s formula was used to calculate the sample size for each subgroup to get a fair representation of the groups. The local government claimed that the registered drivers and traders were 1700 and 600 respectively (Ambrose Agyemang, field engineer, interview 2019), but the latter could be more as field investigation revealed that most traders ignored the registration. With that, we had to decrease the confidence level and increased the margin of error for traders, to at least reduce errors in sample size. We chose a 90% confidence level and 0.10 (or 10%) margin of error for the traders’ sample whereas a 93% confidence level and a 0.07 (7%) margin of error for the drivers’ sample.

The slovin formula is computed as

n = N/(1 + Ne²)

(1)

where n = group sample, N = Total population and e = Error tolerance (margin of error).

Drivers sample = 1700/(1 + (1700 × 0.07²) = 182

Traders sample = 600/(1 + (600 × 0.10²) = 86

However, after an exhaustive field investigation (2019–2022), 160 and 80 drivers and traders positively responded to the interviews, respectively (Table 1). The data on incomes were solicited in Ghana Cedis currency (GH¢).

3.4. Step 2: Analyzing Data Using Theil’s T Statistics

The term decomposability means that the overall inequality of Theil’s T Statistics can be expressed as the sum of within-group and between-group inequalities [33]. Theil’s T Statistics is an index used to measure economic inequality and other economic phenomena [18]. Theil’s T Statistics and the Gini coefficient are most popular inequality measures among others that have been used in empirical studies, but Theil’s T Statistics is preferable because of its decomposability of inequality into three dimensions [33]. Also, the model ensures rich and detailed information on inequality, particularly when there are subgroups to be measured [34].

According to Das and Das [18], Theil’s (1967) entropy measure is defined in the following way;

$$Ty=\sum _{t=1}^{n}Yt\mathit{log}\left(Yt/Pt\right),$$

(2)

First, n is the total population of all groups.

Yt is the share of t in the total incomes, and Pt is t’s share of the total population (t = 1, 2, …, n). The absolute equality in the distribution of income is represented by a situation in which the ratio, Yt/Pt, is unity for all groups. Any deviation of this ratio from unity indicates inequality which Ty in Equation (2) measures. Some terms in Ty take negative values indicating that income shares are less than their respective population shares.

Theil has given proof that Ty ≥ 0 [18].

Ty has a minimum equal to zero, which is attained when for two groups, the income share, and the population share are identical.

The situation of maximum inequality of two groups arises when the income share of one tends to be unity, whereas the income shares of the other tend to be zero. In this case, Ty tends to −log (ph) > 0, which is the measure’s maximum value or inequality.

The decision to include the Theil index, Ty, as a measure of inter-group income inequality is based on considering that the measure in Equation (2) is decomposable.

Now, the decomposition works out as follows:

$$\sum _{t=1}^{n}Yt\mathit{log}\left(\frac{Yt}{Pt}\right)=\sum _{g=1}^{G}Yg log \left(\frac{Yg}{Pg}\right)+\sum _{g=1}^{G}Yg Eg$$

(3)

where

$$Eg=\sum _{k=1}^{ng}Ykglog\left(\frac{Ykg}{Pkg}\right)$$

The dimension on the left-hand side in Equation (3) is the Theil entropy index for overall inequality. This is further decomposed as a summation of several entropy measures appearing on the right-hand side. The first term on the right-hand side is the measure of inequality between groups, with Yg and Pg denoting each group’s income and population shares in the overall total. The first term, therefore, measures inter-group inequality. The second term is the weighted average of inequality levels within groups or intra-group inequality. The inequality within each group is measured by the entropy, e.g., Now, Equation (3) is adopted because it measures the income inequality in three dimensions; overall inequality of traders, inequality between, and within the subgroups.

4. Results and Discussion

This section combined both the results of the study and the discussion part together. That is, it evidenced the results of the impact of resettlement in the Kejetia market redevelopment (phase 1) on the income inequality of the resettled population before, during, and after the resettlement process. In addition, it discusses the determinants and implications of the income changes to the affected population.

4.1. Income Inequality before the Resettlement

Before we get to the main discussion, it is better to take note of the decomposition of traders’ and drivers’ income inequality to understand the form and trend ‘during’ and ‘after’ resettlement stages.

Table 2. Theil’s T Statistics on monthly incomes before resettlement, Source: Field data 2019–2022.

Traders Incomes GH¢	Pop.	Drivers’ Incomes GH¢	Pop.	Theil’s T-Statistics	Between Inequality	T within Inequality	D within Inequality	Overall
0	0	0	0	Total pop. = 240 Total inc. = 271,856	0	0	0	0
4050	7	6100	5	Traders inc. share = 0.3500199	0.0002699	−0.004879	−0.002892	0.097758
4250	1	6500	9			−0.000698	−0.004953
4270	2	6740	12	Traders’ pop. Share = 0.3333333		−0.001397	−0.006380
4470	2	7800	18			−0.001397	−0.007799
4500	2	7960	12	Inc/pop shares = 1.0500596		−0.001397	−0.004996
4560	1	8200	12			−0.000698	−0.004680
4700	5	8500	11	LN of inc./pop shares = 0.0212139		−0.003485	−0.003910
6720	1	8670	23			−0.000612	−0.007709
6790	1	8900	12	Inequality element = 0.0074253		−0.000608	−0.003682
7870	1	8980	7			−0.000519	−0.002077
7950	1	9300	8	Drivers inc. share = 0.6499801		−0.000511	−0.002042
8100	6	10,500	8			−0.002982	−0.000677
8320	6	10,720	7	Driver’s pop. Share = 0.6666667		−0.002850	−0.000356
8350	12	35,200	10			−0.005663	0.065183
9500	13	35,800	6	Inc/pop shares = 0.9749702		−0.004434	0.040356
9860	6					−0.001772
34,900	13			LN of inc/pop shares = −0.011009		0.078018
				Inequality element = −0.007155
Total	80		160			0.044107	0.0533810
						Within groups	0.0974883

H¢ = Ghana cedis, Pop = Population, inc. = income, T = traders, D = Drivers, LN = log of natural logarithm. Interpretation: The Theil index ranges between zero and one, with zero representing an equal distribution and higher values representing a higher level of inequality. Negative index= income shares less than their population shares.

below demonstrates that there was a positive inequality element for traders whiles a negative element for the drivers, indicating that the income shares for traders were bigger than their population shares and vice versa for the drivers. This further indicates that there was more competition for income in the drivers’ group than in the traders’ group. However, the main discussion for the study falls on

Table 3. Theil’s T Statistics on incomes during resettlement, Source: Field data 2019–2022.

Trader’s Incomes GH¢	Pop	Drivers’ Incomes GH¢	Pop	Theils T-Statistic	Between Inequality	T within Inequality	D within Inequality	Overall
0	0	0	0	Total pop. = 240 Total inc. = 3,609,990	0	0	0	0
2950	10	5900	7		0.0001473	−0.0056485	−0.004740	0.13052
3470	9	6400	6	Traders inc. share = 0.321095		−0.0053698	−0.004031
3700	1	6900	1			−0.0006076	−0.000661
3720	1	7280	11	Traders’ pop. Share = 0.333333		−0.0006085	−0.007166
3790	1	7511	1			−0.0006114	−0.000643
3870	1	7610	11	Inc/pop shares = 0.963285		−0.0006146	−0.007044
5250	1	7900	7			−0.0006411	−0.004405
5700	6	8760	12	LN of inc./pop = −0.01624		−0.0038385	−0.007066
6520	5	8980	11			−0.0031317	−0.006345
6720	4	9200	6	Inequality element = −0.00521		−0.0024845	−0.003385
6940	2	9409	6			−0.0012291	−0.003309
7300	2	9500	10	Drivers inc. share = 0.678904		−0.0012041	−0.005459
7330	1	9820	13			−0.0006009	−0.006828
7450	12	10,800	10	Driver’s pop. Share = 0.666666		−0.0071543	−0.004540
35,800	24	17,200	9			0.09349677	0.0021583
		19,670	8	Inc/pop shares = 1.018357			0.0047342
		35,900	18			0.05975156	0.0662132
		41,005	13	LN of the inc/pop = 0.007900			0.0631472
Total	80		160	Inequality element = 0.005364			0.0706220
						Within groups	0.1303736

GH¢ = Ghana cedis, Pop = Population, inc. = income, T = traders, D = Drivers, LN = log of natural logarithm. Interpretation: The Theil index ranges between zero and one, with zero representing an equal distribution and higher values representing a higher level of inequality. Negative index= income shares less than their population shares.

and

Table 4. Theil’s T Statistics on incomes after resettlement, Source: Field data 2019–2022.

Traders’ Incomes GH¢	Pop.	Drivers’ Incomes GH¢	Pop.	Theil’s T-Statistics	Between Inequality	T within Inequality	D within Inequality	Overall Inequality
0	0	0	0	Total pop. = 240 Total inc. = 3,031,109	0	0	0	0
4100	7	3900	7		0.006267	−0.005525	−0.004086	0.126617
6340	3	4500	7	Drivers inc. share = 0.585182		−0.002474	−0.004069
6407	2	4740	12			−0.001647	−0.006924
7510	2	4950	11	Driver’s pop. share = 0.666667		−0.001589	−0.006290
7560	1	5300	7			−0.000792	−0.003922
7705	5	5560	12	Inc/pop shares = 0.877773		−0.003934	−0.006596
7720	1	5980	11			−0.000786	−0.005817
7790	1	6200	6	LN of the inc./pop = −0.05662		−0.000783	−0.003097
7870	1	7200	6			−0.000779	−0.002671
8250	1	7570	10	Inequality element = −0.03313		−0.000761	−0.004137
8760	6	7670	13			−0.004402	−0.005262
9120	14	7830	11	Traders inc. share = 0.414818		−0.009956	−0.004291
11,150	9	7908	1			−0.004936	−0.000382
30,500	27	7960	6	Traders’ pop. share = 0.333333		0.0782257	−0.002266
		7980	1				−0.000375
		8204	8	Inc/pop shares = 1.244454			−0.002831
		25,300	11			0.0398549	0.0329015
		32,600	7	LN of the inc./pop = 0.094979			0.0352673
		35,100	13				0.0753499
Total	80		160	Inequality element = 0.039399			0.0804950
						Within Groups	0.1203499

GH¢ = Ghana cedis, Pop = Population, inc. = income, T = traders D = Drivers, LN = log of natural logarithm. Interpretation: The Theil index ranges between zero and one, with zero representing an equal distribution and higher values representing a higher level of inequality. Negative index= income shares less than their population shares.

, which are explained in detail, representing the conditions during and after resettlement.

4.2. Income Inequality during Resettlement

This stage of the resettlement represents the income situations during their stay at the various relocation sites. Table 3 below evidenced positive and negative forms of inequality elements for the drivers’ and traders’ groups respectively, denoting that though there were inequalities in the said groups, drivers’ income shares were bigger than their population shares whereas traders’ income shares were smaller than their population shares. A situation which is the exact opposite of conditions ‘before resettlement’ in Table 2.

Again, Table 3 demonstrates an increasing trend for traders’ within-inequality and drivers’ within-inequality as compared to their corresponding in Table 2, while a decreasing trend for the inequality between the two groups and their corresponding in Table 2. This means that majority of the individuals within each group will face livelihood problems more than the few successful ones.

4.2.1. Determinants of Income Inequality ‘During the Resettlement’ (At Relocation Sites)

Based on field interviews from 2019–2022, the affected people expressed their views on the factors they perceived to have caused the changes in their income shares during and after resettlement on the market redevelopment. However, these factors are discussed in detail below.

Poor Relocation Sites

Though both the traders and the drivers complained about the conditions on the relocation sites, the traders expressed much dissatisfaction (field interviews 2019–2022). This point was reiterated by one female trader saying;

“We (relocated traders at racecourse) sell here but no buyers. We are like a school with no students” …

(Mary Forkuo, trader, survey 2019)

These traders claimed the relocation sites have no market structures put in place to enhance trading. More so, some of the traders who were relocated to the Abinkyi location, complained about the place being far distant and having low market demand. However, to survive financially, the majority of the drivers routinely go to other non-patronized areas for more passengers, whiles some of the traders abandoned relocation sites to other market-demand areas (field interviews 2019–2022). This induced a disparity in income within the drivers’ and traders’ groups.

Though resettlement may become necessary, those implementing it, whether governments or private developers, need to understand the difference between good and bad practices [35]. Studies like Patel and Mandhyan [36] claimed that relocation sites do not always mean being in far-flung or poor conditions, where life may be unbearable and raise inequity concerns for the affected people.

Lack of Credit Loans

The majority of the traders’ group claimed that the savings and loans companies in the metropolis temporarily suspended all credit loans to them during the displacement and resettlement period (field interviews 2019–2022). One male trader uttered;

“They (Savings and Loans Institutions) agreed to provide us credit facility. Maybe we are still waiting” …

(Paa Solo, trader, survey 2019)

The traders complained that they needed some additional funds to support their business capital because they have incurred some resettlement costs; paying for new business structures on relocated sites and renting new shops elsewhere. On that note, they perceived these issues affected their income levels during their stay at the relocation sites.

However, to overcome the inequality challenge, Delis et al. [37] suggest that credit funding is capable of reducing income inequality. These findings are consistent with the theory of a negative nexus between finance and inequality if access to credit is improved [38].

High Prices of Goods & Services

According to the field survey in 2019–2022, the majority of the relocated traders’ group complained about the high prices of goods limiting them to trade in their desired quantities. However, the few traders who were able to trade in their desired quantities gained more income at the expense of the majority of the traders who couldn’t afford it. In the same vein, the drivers also complained about the situation, as one stated;

“We are driving empty across streets in these hard times. There is no safe store of value for our services. Fuel prices are drowning us” …

(Kabri, driver, survey 2019)

Literature has demonstrated a direct relationship between high prices (inflation) and income inequality; claiming high inflation induces high-income inequality [39,40]. That is, the nominal value of money depreciates and reduces the real income of the beneficiaries indicating low income, hence income inequality [39].

4.3. Income Inequality after Resettlement

Table 4 explains the income inequalities of the two subgroups after resettlement and return to the original site. This stage is the final phase of the resettlement process in the urban market redevelopment. At this point, the trend of income inequality of the two subgroups before, during, and after resettlement, is significant to understand how the resettlement policy has performed on the incomes of the resettled population.

That being the case, the entire resettlement process (Table 2, Table 3 and Table 4) depicts a decreased to an increasing trend of inequality between the traders and the drivers (0.0002699, 0.0001473, 0.006267), and an increased to decreased trend in both ‘within-groups’ (0.0974883, 0.1303736, 0.1203499) and ‘overall-groups ‘inequalities (0.097758, 0.13052, 0.12661). Though the overall inequality after resettlement in Table 4 is slightly decreased (0.12661) as compared to its corresponding (0.13052) in Table 3, it is still higher than the value before resettlement (0.097758) in Table 2.

With that, we conclude that the resettlement process increased income inequality after the redevelopment, which further denotes that livelihood will be a high concern to the majority of the individuals in traders’ and drivers’ groups, particularly the latter.

4.3.1. Determinants of Income Inequality after Resettlement

Limited Place Access

Field observation 2022, re-affirms the drivers’ claim that the final redevelopment design limited their trading function to arrival stops only, hence extremely reducing their income levels. Table 4 confirms the reduction in drivers’ income shares after resettlement and return. Previous studies have demonstrated the relationship between income inequality and place access, particularly the former influencing the latter [41,42]. Also, the present study argues that the limited place access will induce inequity for the drivers’ group.

High Shop Rent

Other traders who were shop users claimed that when their shop owners increased their shop rent after redevelopment, it affected their business capital, hence trading in low quantities and causing low incomes (field interview 2022). Literature affirms the proposition that there is a direct relationship between house rent and income inequality; as house rent rises, income distribution will get worse [43]. However, the vice versa of such a proposition is also supported by literature; that is, high-income inequality worsens rental affordability [44].

Disorganized Marketplace

Some of the traders’ associations such as the Wax print association and Jewelry trade association, claimed that the market organization was poorly managed after redevelopment, therefore making it difficult for customers to find the traders in the new place (field interview 2022). They further claimed that this has caused low market demands for goods, thus reducing their income earnings. However, studies like Pothier & Puy [45] supported this assertion by claiming that changes in the composition of demand affect income distribution in the short run, hence income inequality.

4.4. Implications

At this point, we discussed the effects of the changes in income inequality for the resettled population during and after the resettlement. These are evidenced implications based on field investigations 2019–2022.

4.4.1. Distrust Concerns

Castelfranchi & Falcone [46] relate the concept of trust as a mental state and a complex attitude of one agent towards others about their outcomes relevant to his goal. That is, one’s trust depends on the expectations for others’ performances.

Based on a field investigation of 2019–2022, both the traders’ and drivers’ groups expressed their distrust for resettlement activities now and in the future. With that, most of the traders said they were particularly disappointed with the local authorities for failing to fulfill their promise of giving them temporary shops and good relocation sites.

The drivers also expressed their disappointment with the local authorities for the completed redevelopment project which limited their place of access. That is, before resettlement, the marketplace looks more like a bus terminal than a trading place, but after redevelopment, the place is made up of trading shops.

4.4.2. Indirect Displacement

Indirect displacement is more concerned with the loss of sense of place, and it is less noticeable than direct displacement [20]. That is, people may leave a place voluntarily when they no longer recognize the purpose of their stay. It causes many negative consequences to the displaced population such as loss of jobs, financial stress, emotional discomfort, etc.

Based on field observations in 2022, some of the drivers said a few of their colleagues have relocated to nearby marketplaces for better incomes since the redevelopment project induced low incomes for them. In addition, some of the traders who were relocated to the race-course location voluntarily abandoned the site and moved to other marketplaces due to a lack of market structures and low market demands for their goods. More so, due to the high rent for redeveloped shops and the low market for their goods, some of the low-income traders have shown signs of voluntarily moving to other affordable marketplaces (field/phone interviews, 2022). These situations further raise equity concerns for them.

4.4.3. Personal and Institutional Debt

The majority of the traders’ group claimed that their low incomes during and after resettlement have induced financial stress on them, making them unable to repay their bank loans (field interviews 2019–2022). This has further created a situation where the savings and loans banks have become incapable of paying their depositors or financing daily withdrawals at their premises (field observations 2019–2022). However, the assertion that high levels of income inequality led to higher levels of personal and institutional debt is supported by the literature [47]. More so, Kumhof, et al. [48] suggest that increased inequality is partially responsible for an increase in debts.

4.5. Limitation of the Study

Generally, access to enough data was a challenge during the field survey. That is, it was difficult to college monthly income for a year (12 months), and take an average of 12 to obtain gross monthly income. This was due to the huge financial resources and time required to stay on the field. Instead, monthly income was computed by collecting income data for four consecutive months and dividing it by four. Although the challenge does not affect the study’s results, future research can be improved if annual income data become accessible.

5. Conclusions

The initially stated overarching aim of this research was to examine the impact of resettlement in the Kejetia redevelopment market on income inequality of traders and drivers before, during, and after resettlement, and to discuss their determinants and implications. While recognizing the limitation of the study, we believe we have achieved the study’s aim by answering the three questions of the study; how has income inequality changed across the three resettlement stages; what factors account for the income changes; and what are the implications for the resettled population?

Theil’s T Statistics evidenced complex income inequalities of the resettled population both in form and trends across the entire resettlement process, using its three-dimensional features; ‘within-group inequality,’ ‘between-group inequality,’ and ‘overall inequality.’

In terms of the form, the study findings highlighted both positive and negative inequality elements at each resettlement stage for traders and drivers, denoting that positive inequality has its income shares bigger than the population shares, and vice versa for the latter (negative inequality). However, for the trend in inequality across the three resettlement stages, the study claimed that there was a decrease to an increasing trend in ‘between-groups inequalities’ and an increased to slightly decreased trend in both ‘within-groups’ and ‘overall groups’ inequalities.

The determinants for such decomposition ‘during resettlement’ include poor relocation sites, lack of credit loans, and high prices for goods whereas that of ‘after resettlement enlisted limited place access, high shop rent, and market disorganization. The implications for the resettled population included indirect displacement, distrust concerns, and personal and institutional debt. Generally, the study concludes that the resettlement project did not only induce unsatisfactory incomes and inequality, but equity concerns for traders’ and drivers’ groups, particularly the latter.

The key implication of the research is the importance of influencing scholars and policy practitioners to give attention to the impacts of resettlement in urban market redevelopment on income inequality of the resettled population.

The study recommends future research to have more methodological work on collecting annual income data from the field to improve research. Also, the study encourages constructing the same research in a new context or location to weave together present thought and data with future knowledge.

By and large, the study contributes to the literature on urban redevelopment-induced resettlement, especially urban market redevelopment.