Abstract
This article investigates whether urban social movement participation influences post-eviction resettlement success. Pre- and post-eviction interviews were conducted with sixty-four slum dwellers from Dar es Salaam, Tanzania, including members of the Tanzania Federation of the Urban Poor (TFUP). The majority of interviewees reported improved post-eviction housing but adverse employment impacts. TFUP membership was negatively associated with employment outcomes, particularly for property owners. Expecting TFUP to secure housing for them, members delayed finding accommodation. This led to resettlement farther from their former homes and negatively affected employment. Women were especially vulnerable following eviction, with their post-eviction pay falling because of the nature and location of their pre-eviction work.
Introduction
Forced evictions and involuntary resettlement of individuals, families and communities rank among the most widespread human rights violations in the world: in 2007-2008, 4.31 million people were affected by implemented or threatened forced evictions (COHRE 2009). In the developing world, in particular, there is a hope that grassroots mobilization can serve as a means for marginalized groups to address development-related challenges—such as forced eviction and involuntary resettlement—in the face of limited state resources or willingness to provide for their needs (Zaidi 1999). This article investigates whether participation in an urban social movement—specifically the Tanzania Federation of the Urban Poor (TFUP)—influences post-eviction resettlement success. It draws on detailed pre- and post-eviction interviews with a sample of sixty-four slum dwellers from Kurasini, Dar es Salaam, Tanzania. 1 The article first describes evictees’ resettlement outcomes with respect to housing, employment, income, and resettlement distance. Drawing on qualitative and quantitative evaluations of interviewees’ resettlement accounts, it then examines whether social movement participation influenced post-eviction resettlement success.
Despite the large number of urban evictions taking place worldwide, there has been limited research on involuntary urban resettlement, especially in the developing world (Scudder 2005; Olds, Bunnell, and Leckie 2002). By following a group of extremely poor evictees (with an average household income of 2.93 U.S. dollars per day 2 ) through the eviction process—from before eviction to after resettlement—this article contributes to literatures on both the dynamics of social movement participation and the consequences of evictions. While centered on a single case study in Tanzania, the findings are applicable in a wide variety of similar contexts. The economic and political forces currently shaping Dar es Salaam and around which TFUP mobilizes—including increasing demand for the marginal land on which the poor live—also influence other cities in Africa and elsewhere in the developing world (Smets 2002). The form of social movement mobilization adopted by slum dwellers in this case study—the enumeration of to-be-evicted households by members of a slum dweller movement—is increasingly employed by similar groups around the world (Weru 2004). Given that more than 70 percent of the urban population in most Sub-Saharan African countries reside in slums (Izutsu and Tsutsumi 2009), many of which face the threat of eviction, the study’s findings are widely relevant.
This article focuses on a case study of mobilization around evictions in Dar es Salaam’s Kurasini ward, which lies adjacent to the city’s port and is home to more than 34,500 people (Figure 1). Plans to “reorganise the land use pattern [of Kurasini] to provide more land for storage of transit goods” have circulated since 2001, when the Ministry of Lands, Housing and Human Settlements Development published the Kurasini Area Redevelopment Plan (Government of Tanzania 2001). In October 2007, the first wave of evictees from Kurasini was ordered to leave their homes. Following common Tanzanian practice, property owners received some compensation on eviction for lost property while renters received none. 3 Rather than simply representing an example of eminent domain, the eviction in Kurasini falls under the rubric of a forced eviction because the process by which the eviction was carried out lacked transparency and was enforced under threat of physical force by government officials. A number of factors highlight the forced nature of the eviction: compensation values were not disclosed until the moments immediately prior to eviction; residents were made to confirm acceptance of their compensation packages before viewing the values on their individual checks; compensation values varied widely between households; compensation was paid in the presence of bulldozers, armed police, and police dogs; and physical altercations took place between the police and individuals who questioned the eviction process (Kisembo 2008).
Locations of Tanzania (top left), Dar es Salaam (top right), and Kurasini ward (bottom)
TFUP, Tanzania’s national slum dweller movement, is built on a nationwide network of community savings groups and has seventeen hundred members in thirty savings groups across the country (Ndezi 2009). Prior to eviction, Kurasini ward was home to seven of these groups, accounting for approximately three hundred members among them. 5 While ostensibly organized around the issue of daily savings and the granting of small loans to members, TFUP also has strong movement dimensions, with the savings groups serving as the basis for more political and risky forms of mobilization around local land, housing, and sanitation issues (Hooper 2010). TFUP is associated with Slum Dwellers International (SDI), a global network of slum dweller movements, and works with the Centre for Community Initiatives (CCI), a Tanzanian partner NGO that provides capacity-building support. TFUP is a remarkably horizontal organization, with most decisions being based in discussions at savings group meetings or larger assemblies. The observation that Appadurai (2001, 31) makes about SDI more broadly also applies to TFUP: “[it] cultivates a highly transparent, non-hierarchical, anti-bureaucratic and anti-technocratic organizational style.” A core of organizers that is particularly involved in TFUP leadership activities has emerged through self-selection from the wider membership and develops strategy with the assistance of CCI’s two staff members. The organizers are slum residents and, in this study, faced imminent eviction.
The principal pre-eviction mobilization effort undertaken by TFUP members in Kurasini was an enumeration of to-be-evicted households. Enumerations consist of a population census and comprehensive mapping of plots and households (Weru 2004). Enumerations demonstrate that marginalized communities have the capacity for self organization, generate data that give a tangible identity to slum residents, and serve as the basis for lobbying and advocacy. Accepting the Kurasini eviction would take place, TFUP used the data generated through the enumeration to lobby municipal and national agencies responsible for land administration for a grant of land on which evictees could be resettled prior to eviction. 6 While no direct promises were made to members by movement organizers concerning resettlement, the idea that evicted TFUP members would receive a grant of land became well established in evictees’ everyday discussions and planning, not least as a result of brainstorming exercises in which TFUP members worked with local architects to visualize their post-eviction homes.
TFUP organizers focused their lobbying exclusively on the housing-related consequences of the eviction. This strategy was based on an assumption, similar to that made by policy makers and community leaders in other evictions, that the bulk of negative impacts arising from displacement would be related to housing (Mejia 1999). Movement organizers expected that it would be very difficult for evictees to secure satisfactory post-eviction housing and believed a grant of land for resettlement would help offset these adverse housing impacts. 7 Six months after eviction, these efforts still had been unsuccessful and residents—including TFUP members and non-members—had received no grant of land or group resettlement assistance. As a result, all residents were responsible for individually identifying post-eviction housing options on their own and carrying out their own resettlement.
While TFUP may be smaller in scale and scope than the kinds of movements that have dominated social movement scholarship in Europe and the United States, it is nonetheless representative of the increasingly wide range of relatively small, yet transnationally connected, movements taking root across the developing world (Batliwala 2002). Tarrow (1998, 4) defines a social movement as “collective challenges, based on common purposes and social solidarities, in sustained interaction with elites, opponents, and authorities.” By engaging in a sustained, public mobilization effort around the Kurasini eviction and by consistently demanding government authorities provide land for resettlement, TFUP organizers and members made a strong collective claim on the Tanzanian state. In this way, TFUP satisfies the aforementioned definitional requirements of being a social movement.
Data and Methodology
Pre- and post-eviction interviews were conducted with a sample of sixty-four slum dwellers in Kurasini between July 2007 and July 2008. All individuals in the sample were interviewed before and six months after eviction. Interviews were conducted in English and Swahili by the first author and a research assistant. Interviewees were selected by virtue of being in the first cohort of evictees and were evicted between October and December 2007. All TFUP members from the first evictee cohort known to movement organizers were interviewed. Non-members were identified through a call for interviewees circulated in Kurasini, and respondents who lived closest to a randomly chosen interview site were selected as interviewees. 8 Only one evictee per household was interviewed. Interviewees were not paid for their participation, but were compensated for travel to the interview venue. Pre- and post-eviction interviews lasted approximately two hours and, although structured around a common set of questions, were open ended.
The group of sixty-four interviewees consisted of forty-three TFUP members and twenty-one non-members (see Table 1 and Figure 2). The sample was weighted toward TFUP members because the larger research project, of which this article is one part, focused on the dynamics of movement membership and more time was allocated to interviewing members than non-members. The evictee sample consisted of 64.1 percent owners and 35.9 percent renters. With respect to gender, the sample included 60.9 percent women and 39.1 percent men. The dominance of females in the sample in part reflects the predominance of women in the TFUP movement, which, like many African social movement organizations, began as a women’s movement. Given the relatively small sample size and recognizing that care must be taken in interpreting the results of statistical tests involving few observations, the article uses both qualitative and quantitative analytical techniques.
Composition of the Study Sample (n = 64)
Evictee sample breakdown in terms of TFUP membership, owner/renter status, and gender (n = 64)
The analysis proceeds in two parts. The first draws on an analysis of evictees’ interview transcripts to identify key patterns in post-eviction resettlement success. 9 The transcripts reflect interviewees’ perspectives on eviction and resettlement, and it is important to recall that they are self-reported accounts. Resettlement success was assessed in terms of housing, employment, household income, and resettlement distance. Post- versus pre-eviction housing was evaluated by interviewees on a scale of worse/same/improved in terms of physical setting, 10 security of tenure, 11 and cost. Post- versus pre-eviction employment (also evaluated according to worse/same/improved) was assessed in terms of travel time to work, hours worked, status, 12 security, 13 and pay. Post- versus pre-eviction household income was gauged in terms of percentage change in income. Resettlement distance was evaluated in terms of kilometers between the evictees’ post- and pre-eviction places of residence. The second stage of the analysis undertakes a more detailed examination of evictees’ resettlement accounts. These accounts were evaluated qualitatively by identifying key themes and issues across interview transcripts. They were also evaluated quantitatively by tabulating and coding responses to determine whether patterns identified in evictees’ accounts were statistically significant.
Patterns of Resettlement Success
Post-Eviction Housing
For all measures of post-eviction housing, a majority of interviewed evictees reported their housing situation as improved (ranging from 54.7 percent, for housing cost, to 87.5 percent for housing location) (Table 2). This is striking since TFUP’s mobilization efforts focused on the adverse housing-related impacts of the eviction. Contrary to TFUP movement organizers’ expectations, and despite not gaining a grant of land for resettlement at the time post-eviction interviews were conducted, post-eviction housing conditions worsened for only a small minority of evictees.
Evictee Assessments of Post- versus Pre-Eviction Housing, Employment, and Household Income
Note: Values represent percentage of interviewees answering in each category.
Post-Eviction Employment
For all measures of post-eviction employment, a minority of evictees reported improved conditions, ranging from a low of 1.8 percent for travel time to work to a high of 22.8 percent for hours of work (Table 2). The number of interviewees who provided responses concerning employment was less than the full sample, since evictees who were unemployed prior to and after eviction did not answer some employment-related questions. For three of the five measures, a majority of evictees reported their post-eviction employment situation as the same as that before eviction, while for the remaining two—travel time and pay—a majority reported their post-eviction status as worse. The most negative change concerned travel time to work, with 78.9 percent of evictees reporting worsened conditions after resettlement. This is an important finding since an increase in travel time to work can adversely affect other aspects of well-being, for example through reduction of time available to work (Stutzer and Frey 2008). 14 Individual pay also suffered, with nearly half (49.1 percent) reporting worsened post-eviction pay.
Post-Eviction Household Income
The majority of interviewed evictees (59.4 percent) reported their household income as improved following eviction, compared with their pre-eviction circumstances (Table 2). The average post-eviction household income of evictees was 169,000 TSh/month, compared with 118,000 TSh/month prior to eviction, a 43.2 percent increase. On first glance, these results appear to conflict with those from the previous section, which showed that only a small minority of evictees (14.0 percent) reported increased individual pay following eviction and that 49.1 percent reported a fall in pay. This seeming contradiction will be addressed in greater detail later in the article’s analysis of evictees’ resettlement accounts.
Post-Eviction Resettlement Distances
Post-eviction owners resettled significantly farther than post-eviction renters from their former homes (Student’s t-test, p = 0.000) (Figure 3). 15 On average, post-eviction owners resettled 4.5 km from their former homes, compared with 1.3 km for post-eviction renters. The resettlement locations of post-eviction owners also exhibited considerably more variation than those of renters. Post-eviction owners had a range of 17.0 km in resettlement locations, while post-eviction renters had a range of only 3.7 km. Finally, among post-eviction owners, but not renters, TFUP members resettled significantly farther from their former homes than non-members (Student’s t-test, p = 0.006). Post-eviction owners who were TFUP members resettled an average of 6.0 km from their former homes, compared with 2.3 km for owners who were non-members.
Evictee resettlement distances, in average kilometers from place of pre-eviction residence
Analysis of Resettlement Accounts
This section examines evictees’ narrative accounts of resettlement to better understand the ways in which social movement participation influenced post-eviction resettlement success. It also addresses an apparent contradiction identified in the description of resettlement outcomes, which showed that post-eviction household incomes increased for the majority while only a minority reported improved individual pay. Given the relatively small sample size and the challenges of drawing inferences therefrom, both quantitative and qualitative analytic methods were used to develop as robust a characterization as possible of patterns in the data. That said, the sample size is such that results should be viewed as strongly suggestive rather than definitive.
TFUP Membership and Resettlement Success
Member expectations and delayed housing search
Evictees’ narratives indicate that TFUP membership adversely affected resettlement outcomes by raising their resettlement expectations and negatively influencing their strategies for securing post-eviction housing. TFUP members delayed securing their own post-eviction homes in the expectation that the TFUP-led effort to gain a grant of land for resettlement would be successful. In a comment representative of the attitude of evicted TFUP members, a member noted prior to eviction: “Through the federation I can get a house in the near future. The federation is trying to address this problem of housing.” This focus on the future—but as yet unrealized—provision of housing by the movement was also emphasized before the eviction by a TFUP member: “We haven’t received help with housing yet, but I am still expecting to get a house and land from the federation.”
TFUP members made an explicit connection between an expectation of housing from the movement, their own delayed housing search and, ultimately, as a result of the delay, having to resettle farther than expected from their former residence. While both post-eviction owners and renters who were TFUP members reported a similar pattern of delayed housing search, this strategy proved especially challenging for post-eviction owners. Given their resettlement at significantly greater distances from their former homes, a relatively distant home for a post-eviction owner was significantly farther from Kurasini than that for a post-eviction renter who likewise delayed their housing search. As reported earlier, interviews revealed that post-eviction owners who were TFUP members resettled an average of 6.0 km from their former homes, compared with 2.3 km for owners who were non-members.
Highlighting the connection between delayed housing search and resettlement distance, particularly for post- eviction owners, an evicted member and post-eviction owner stated: “We thought we would get housing from the movement. I have been waiting, but nothing has come. Now I’ve had a difficult time to find my own house and have moved far away [from Kurasini].” The direct connection between delayed search and greater resettlement distance was emphasized by other TFUP members, including a post-eviction owner, who said: “Because I waited, I now live very far away. I didn’t want to spend my money if we would get good housing, but now I will have to live far from Kurasini.” This sentiment was similar to that expressed by another TFUP member and post-eviction owner, “We were really expecting a house or land, but we couldn’t wait so long. I already had to find a place that is very far because too many people are seeking plots due to the eviction.”
These narrative sentiments concerning the expectations of TFUP members are substantiated by additional data drawn from evictee interviews. In post-eviction interviews, evictees were asked to retrospectively report the factors that they believed before and during the eviction would help them secure better post-eviction housing. Fifty-two percent of interviewees cited one of three TFUP-affiliated organizations—the TFUP movement as a whole, TFUP’s savings groups, and the local TFUP-affiliated partner NGO, CCI—as the factor they thought would be most important in securing better post- eviction housing. This made the TFUP movement, broadly construed, the single most important perceived source of improved post-eviction housing. The perception that the TFUP movement could be an important source of potential housing assistance likely reflects the movement’s strategic and rhetorical focus on the adverse housing impacts of the eviction.
TFUP members were significantly more likely than non-members to consider TFUP an important factor in potentially securing better housing (chi-square test, p = 0.001). Of members, 77.5 percent mentioned they believed the TFUP movement could help them secure better housing, compared with 0.0 percent of non-members. While members were more likely to consider TFUP a source of better post-eviction housing, non-members were more likely to look to the government for assistance. Non-members were significantly more likely than TFUP members to cite “government” as a possible source of improved housing (Fisher’s exact test, p = 0.044), with 40.0 percent of non-members citing government as compared with 13.5 percent of members.
Despite more than three-quarters of TFUP members stating prior to eviction they had considered the movement and its affiliated organizations important sources of improved post-eviction housing, evictees’ ex post assessments of the movement’s actual role in securing post-eviction housing were decidedly different. Six months after eviction, no evictees mentioned TFUP as a factor that helped them secure their actual post-eviction housing. When asked explicitly whether they had received “any help from the TFUP movement or its savings groups in finding [their] new home,” 100 percent responded in the negative. The contrast between the levels of housing assistance expected and received by members helps explain their delayed efforts to secure housing and their relatively greater distance of resettlement.
Delayed housing search and employment impacts
TFUP members’ delayed action on housing had spin-off effects on employment after eviction. The decision to defer their housing search was particularly problematic for evictees who resettled as owners, because of their resettlement farther from Kurasini than renters. Even if post-eviction owners faced few negative impacts with respect to housing, their employment prospects were adversely affected by their resettlement at relatively distant sites, with many owners remarking that they either struggled to travel to their former jobs or develop new jobs far from their social networks. An evicted TFUP member and post-eviction owner stated, “By waiting so long, now I live very far away—15 km. It is very bad for my business.” Another evicted TFUP member and post-eviction owner highlighted the connection between delayed action on housing and negative employment impacts: “I expected housing and was waiting. I waited too long and found a poor site. Now my work suffers.” And yet another evicted member and post-eviction owner made the connection even more explicitly: “Due to the eviction, people have had to move very far away and this has affected employment. They must travel to the city or find new work far away; both are not good options. They should have found a home closer to Kurasini, but we were too slow and all the good places were gone. I thought I would get a home from the federation, but I had to fight on my own.”
The narrative accounts of evictees coincide with a quantitative analysis of interview data, which shows that post-eviction owners who reported unimproved post-eviction employment hours, status, and pay resettled significantly farther from their former homes than those who reported improved conditions (Figure 4).
Relationship between resettlement distance and employment outcomes for post-eviction owners
Post-Eviction Changes in Individual Pay versus Household Income
A close examination of evictees’ interview transcripts shows that men’s and women’s individual pay responded differently after eviction. This likely accounts for the complex relationship between post-eviction household income and individual pay. As described earlier, the majority of households (59.4 percent) reported improved post-eviction household income, but only 14.0 percent of evictees reported improved individual pay. While the majority of both male and female interviewees reported improved post-eviction household incomes, there is a statistically significant difference in post-eviction pay for men and women (chi-square test, p = 0.004). The majority of men (76.2 percent) reported similar or improved post-eviction pay while the majority of women reported worsened post-eviction pay (63.9 percent). Furthermore, the majority of interviewees in this study (62.5 percent) were women. This strongly suggests that the stability, and in some cases growth, of men’s post-eviction pay drove the increase in post-eviction household income, while the decline in individual pay for the majority of interviewees reflects the lower post-eviction pay reported by women (Figure 5). Since the majority of evictees (73.4 percent) were in marital relationships, this indicates a complex financial dynamic within households.
Changes in post-eviction household incomes and male versus female individual pay Note: HH = Household.
The nature and location of men and women’s pre-eviction work likely exerted a strong influence on the direction of changes in their individual pay following eviction. Prior to eviction, women were significantly more likely than men to work in the eviction zone (Fisher’s exact test, p = 0.005). Eighty-eight percent of women worked in the eviction zone, compared with 47.1 percent of men. As a result, women interviewees were more likely than men to have to relocate both their homes and jobs after eviction. Women working in the eviction zone were also significantly more likely than men to be engaged in biashara, the Swahili term for petty or street trading (chi-square test, p = 0.000). Of women working in the eviction zone prior to eviction, 96.4 percent were employed in biashara. None of the men working in the eviction zone were engaged in biashara.
Biashara—as well as women’s informal work more generally—is particularly dependent on social networks and close social relationships that are difficult to reestablish after resettlement. Prior research suggests that disruption of these networks would have been particularly challenging for women’s post-eviction work prospects (Parasuraman 1999). As evidence of this, women who reported worsened pay after resettlement relocated significantly farther from their former homes in Kurasini than those that did not (Student’s t-test, p = 0.037). Women who reported worsened pay (n = 23) resettled an average of 3.17 km from their former homes, while women who reported the same or improved pay (n = 13) resettled an average of 1.3 km from their former homes.
Evictee assessments of the challenges faced due to eviction provide further evidence that women’s social networks were disrupted by moving farther from their former homes and that this was problematic for post-eviction employment. Women who cited “finding new friends and acquaintances” as a major challenge of the eviction resettled significantly farther from their former homes than those who did not mention this challenge (Student’s t-test, p = 0.000). The narrative accounts of female biashara pracitioners also support the argument that relocating farther from Kurasini was particularly challenging for reestablishing their business networks. For example, an evicted woman who was engaged in biashara before eviction said: “To live in a new settlement is very difficult at first. I used to sell juices and fruits at my home place. But the business is not doing well at my new home because people need to know and understand me.” A similar sentiment was expressed by another female biashara practitioner, who stated: “To cope with the new environment was a great challenge because I had no people who knew me—this affects me economically.”
Discussion
Results presented here reinforce key findings from previous research on involuntary resettlement and extend them to an urban eviction in an African context. For example, Mejia (1999, 176) contends there is “a prevailing belief that resettlement of the urban poor does not seriously disrupt economic activities,” and this has led to a predominance of attention on post-eviction housing concerns rather than employment. Scudder (2005, 120) likewise notes that the “lack of attention to economic development is . . . especially glaring.” In his review of dam-induced resettlement, Scudder (2005) found that in addition to the more expected challenges of landlessness, joblessness was a problem in 80 percent of cases. This supports a key finding of our study of Kurasini: employment challenges were a significant, and unappreciated, problem in the resettlement, while attention and expectations of TFUP organizers and members centered primarily on housing.
The results also confirm key findings from prior research concerning the relationship between resettlement, employment, and gender. Parasuraman (1999) notes that women are especially vulnerable because they are more dependent on kinship and other social ties. Similarly, notwithstanding an increase in household incomes in our study, women’s individual pay worsened following eviction as they were more likely to have been previously employed in the eviction zone and to be engaged in work that relied on close social networks. In arguing for further research on the relationship between resettlement location and informal employment, Mejia (1999, 177) contends the belief that economic factors are a secondary consideration in resettlement “persists, in part, because of the scarcity of data on the correlation between economic activities and place of residence.” The findings from this case provide a link in understanding the relationship between the pursuit of certain economic activities, particularly biashara, and the adverse impacts of resettlement.
To date, the literature has largely focused on the positive aspects of social movement participation and has neglected potential downsides (Portes and Landolt 1996). At the same time, it is clear that some social movements either do not achieve the results they intend or produce unanticipated results (Giugni 1998). Tilly (1999, 268) remarks that the unanticipated consequences of social movements “far surpass the explicit demands made by activists in the course of social movements, and sometimes negates them.” The lack of attention to negative and unanticipated impacts of movement mobilization may be due to a sentiment that even where social movements fail to achieve explicit goals, they often succeed in producing other positive societal changes. As Tarrow (1998, 2) states, with reference to movements of the latter half of the twentieth century, “they have often succeeded, but even when they failed, their actions set in motion important political, cultural and international changes.” While this may be true at the societal level, what of the impact of unsuccessful mobilization efforts on participants themselves? While movements may have spin-offs that ultimately benefit, and even transform society, what happens to individual participants who do not achieve their goals in the short term?
The results of this study point to a need for closer examination of unsuccessful mobilization efforts and for greater attention to be paid to the expectations of participants. As this study shows, where expectations are unrealistic, they may ultimately prevent more pragmatic action to protect participants’ self-interests. Theory on “escalation of commitment” helps to explain why those involved may pin their hopes on a mobilization effort even when the odds appear stacked against success. In this case, escalation of commitment was observed in a persistent belief that evictees would receive housing from TFUP.
Escalation theory applies to situations “where losses have been suffered, where there is an opportunity to persist or withdraw, and where the consequences of these actions are uncertain” (Staw 1997, 192). The theory holds that individuals can become locked in losing courses of action and throw good money (or time or other resources) after bad. 16 Prior research has shown that optimism, the illusion of control, and self-justification can provide explanations for escalation behavior (Taylor 1989; Northcraft and Wolf 1984; Brockner 1992). Determining if these psychological factors play a role in influencing evictee decision making around resettlement is likely to be a fruitful avenue for further research. Irrespective of the precise psychological drivers of escalation behavior, the present results suggest that in charged circumstances such as involuntary resettlement, where much is at risk and where much is expected from a particular course of action, there is considerable risk of entrapment in escalation scenarios.
Conclusion
Across all measures of post-eviction housing—physical setting, tenure security, and cost—a majority of evictees in this study reported improved conditions. In contrast, across all employment measures, a majority of evictees reported conditions worse than or similar to those before eviction. These results are surprising since TFUP’s pre-eviction mobilization strategy focused on housing issues and not on employment and suggest that the movement may have been better served by focusing their resettlement strategy around employment-related consequences of eviction, rather than on adverse housing impacts.
Evictees’ narratives indicate that TFUP membership negatively affected resettlement outcomes by raising members’ resettlement expectations and adversely influencing their strategies for securing post-eviction housing. In interviews conducted six months after the eviction, 100 percent of evictees reported receiving no housing assistance from the movement. TFUP members reported that they intentionally delayed action in securing post-eviction housing because they expected land and housing to be provided. The delay forced members to resettle relatively farther from their former homes than non-members. This strategy was particularly problematic for members who were post-eviction owners, since—given their resettlement in locations farther from Kurasini than post-eviction renters—securing housing later was likely to leave them living especially far from their former homes. Delayed action on housing led TFUP members, particularly those who were post-eviction owners, to experience more negative outcomes with respect to post-eviction employment. These evictees either had to commute long distances to jobs near their former homes or find new forms of employment close to their new places of residence.
While the majority of evictees reported increased household incomes after resettlement, women’s individual pay fell. Prior to eviction, women were significantly more likely than men to work in the eviction zone and to work in biashara, a form of employment especially dependent on tight social networks. As women moved farther from their homes, they experienced more severe disruptions to their social networks and were more likely to report worsened post-eviction pay. This suggests that women are particularly vulnerable in evictions, especially since the negative impacts they face may be disguised by aggregate improvements at the household level.
Findings from Kurasini confirm key results from the limited prior research on involuntary urban resettlement in the developing world. They reinforce the generally underappreciated impact of evictions on the employment prospects of evictees and for women in particular, and indicate that these findings may hold more widely, particularly in the rapidly growing cities of Sub-Saharan Africa. Theory on escalation of commitment may help to explain why TFUP members delayed their housing search in the expectation that the movement would help them secure new homes and may serve as a fruitful basis for further research.
Results from this research pose considerable practical challenges to urban planners, policy makers, and movement organizers. They show that the expectations of movement members can be strongly shaped by the rhetorical and strategic focus of a mobilization effort. When mobilization does not produce expected results, members can be left facing worse circumstances than they might otherwise have encountered. For urban planners, the results suggest that grassroots mobilization alone may be insufficient to provide marginalized groups with an effective means of confronting and coping with rapid urban change. Despite the growth in civil society groups in cities in the developing world, there remains a strong need for planners to ensure that the needs of marginalized groups are considered in formal planning procedures and that proactive efforts are made to support communities in their efforts to cope with highly disruptive events, such as evictions. Furthermore, the results highlight several distinct cleavages among evictees that must be carefully considered in resettlement efforts. The first involves post-eviction owners and renters, who here faced considerably different impacts, based on their different resettlement locations. The second involves male and female evictees. As this research shows, women are especially vulnerable because of the nature and location of their pre-eviction work.
Footnotes
Acknowledgements
Thanks are extended to the members of the Tanzania Federation of the Urban Poor (TFUP) and staff of the Center for Community Initiatives (CCI) who provided valuable assistance in the field.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by funds provided to Michael Hooper through a Doctoral Fellowship from the Canadian Social Sciences and Humanities Research Council (SSHRC), a Goldsmith Fellowship from the Stanford Center on International Conflict and Negotiation (SCICN), an International Research Grant from the American Planning Association, and grants from the Stanford’s Center for Philanthropy and Civil Society and Emmett Interdisciplinary Program in Environment and Resources. UPS Foundation support for Leonard Ortolano is also gratefully acknowledged.