Land titling, perceived tenure security, and housing

ARTICLE IN PRESS Habitat International xxx (2009) 1–8

Contents lists available at ScienceDirect

Habitat International journal homepage: www.elsevier.com/locate/habitatint

Land titling, perceived tenure security, and housing consolidation in the kampongs of Bandung, Indonesia Gustaaf Reerink a, *, Jean-Louis van Gelder b a b

Van Vollenhoven Institute, Faculty of Law, Leiden University, P.O. Box 9520, 2300 RA Leiden, The Netherlands Netherlands Institute for the Study of Crime and Law Enforcement, P.O. Box 71304, 1008 BH Amsterdam, The Netherlands

a b s t r a c t Keywords: Land titling Property rights Perceived tenure security Housing investment Indonesia Bandung

In an effort to improve slums and reduce poverty, many developing countries have embarked on mass land titling programs. Yet empirical evidence supporting their assumed beneficial effects has not been readily forthcoming. Following the implementation of mass titling programs in Indonesia, this paper empirically examines the relationships between land titling, perceptions of tenure security, and housing consolidation in seven kampongs in the city of Bandung. It presents evidence that titling indeed contributed–albeit modestly–to people’s perceptions of tenure security, and that both land titling and perceived tenure security enhance housing consolidation; but also that the relationship is more problematic than is often assumed. The results put into doubt whether titling programs in urban Indonesia are actually benefiting the right group, as those who would be best served by the programs can rarely participate in them. Ó 2009 Elsevier Ltd. All rights reserved.

Introduction In recent years, there has been renewed attention for land tenure security as an instrument for slum development and poverty reduction. In developing countries worldwide, the dominant approach to attain this has been to formalise what some call ‘extra-legal’ (e.g. De Soto, 2000) land tenure of the urban poor through mass titling programs. It is assumed that this not only affords people legal protection against involuntary removal from their land by the state or third parties, but also generates an economic multiplier effect that inter alia includes an increased willingness on the part of slum dwellers to invest in their land and housing, ‘enabled’ land markets, and improved access to credit from banks (De Soto, 2000; World Bank, 2003). While titling programs have been and continue to be implemented in developing countries worldwide, relatively little empirical research has been conducted into their effects in urban areas (Payne, Durand-Lasserve, & Rakodi, 2007) and to the extent that it has, the evidence produced has been equivocal. Many of its assumed effects, such as increased access to credit, have failed to materialize (e.g. Calderoń, 2004; Durand-Lasserve & Selod, 2007; Pamuk, 2000; Van Gelder, 2009). Moreover, titling programs have been said to potentially generate unintended side effects. Costs associated with

* Corresponding author. Tel.: þ31 71 5277260; fax: þ31 71 5277670. E-mail address: [email protected] (G. Reerink).

formalisation, such as taxes and registration fees, for example, can lead to difficulties when poor residents are unable to keep up with their payment (Sjaastad & Cousins, 2008). Additionally, land titling has been said to potentially carry detrimental effects such as the disruption of community life and social networks (Bromley, 2008; Payne, 1997; Von Benda-Beckmann, 2003). Not surprisingly, the titling approach has met with substantial criticism. An important point regards the assumption of a dichotomy between the legal and the ‘extra-legal’. Standard land titling approaches term property with registered title legal, while lumping all property that lacks it in the extra-legal category. However, it has repeatedly been argued that the borderline between the formal and the informal in cities in developing countries is usually more blurred than clear (Azuela, 1987; DurandLasserve, 2006; Payne, 1997, 2002). Instead of using a dichotomy, it would hence be more accurate to view tenure categories as a continuum along which tenure categories vary in terms of their degree of legality (Durand-Lasserve & Selod, 2007; Fernandes & Varley, 1998; Payne, 1997, 2001; Razzaz, 1993; Varley, 2002). Few studies however, have empirically examined this notion. Furthermore, titling approaches tend to downplay or even ignore the fact that ‘extra-legal’ tenure can offer significant de facto protection against involuntary removal (Durand-Lasserve & Selod, 2007; Payne, 2002; Payne et al., 2007). Such security depends on factors such as political commitments or administrative practices, the duration of occupation, the size of the settlement, the degree of

0197-3975/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.habitatint.2009.07.002

Please cite this article in press as: Reerink, G., van Gelder, J.-L., Land titling, perceived tenure security, and housing consolidation in the kampongs of Bandung, Indonesia, Habitat International (2009), doi:10.1016/j.habitatint.2009.07.002

ARTICLE IN PRESS 2

G. Reerink, J.-L. van Gelder / Habitat International xxx (2009) 1–8

community organization, and the support that residents can obtain from civil society groups (Durand-Lasserve, 2006; Durand-Lasserve & Royston, 2002; Gilbert, 2002; Kundu, 2004; Payne, 1997). Another main line of criticism with respect to land titling approaches regards the assumption that dwellings in slums are in poor condition primarily because residents lack a property title, and are therefore unwilling to invest in their homes (Razzaz, 1993). Yet, people have been found to improve their houses also in spite of not having legal title (Gilbert, 1994; Van Gelder, 2007; Varley, 1987). Legality of tenure may therefore not be a necessary precondition for establishing security of tenure, as security depends less on the exact legal status and more on occupants’ perceptions of the probability of eviction and demolition (e.g. Broegaard, 2005; Gilbert, 2002, 1994; Payne, 2002, 2001; Razzaz, 1993; Turner, 1976; Varley, 1987). When perceptions of tenure security ultimately form the basis upon which the dweller takes decisions and acts (see also Broegaard, 2005; Jansen & Roquas, 1998; Sjaastad & Bromley, 2000), the relevant question is to what extent land titling programs are indeed an effective approach to influence this perception. In other words, do people actually consolidate faster when having title? If they do not, alternatives to titling programs may be more productive in generating the desired effects. The present paper addresses this question with respect to kampongs in Bandung.

Land titling programs in Bandung, Indonesia Indonesia is one of the countries that has been implementing mass titling programs on a large-scale basis, both in rural and urban areas. In 1981 the Indonesian government initiated the National Agrarian Operation Project (Proyek Operasi Nasional Agraria or PRONA), which aimed to increase tenure security for economically vulnerable residents by allocating property titles and the resolution of land disputes.1 The government also believed that this measure would contribute to the socio-political stabilisation of Indonesian society.2 Around 1988 the government stepped up its land titling efforts, and the number of newly titled parcels grew to about 1 million per year across Indonesia. However, since the total number of parcels increased even faster than before, the government was unable to catch up (World Bank, 1994) and the need for a more ambitious approach emerged. In 1994 the Indonesian government and the World Bank started the five-year Land Administration Project (LAP). Besides land titling, another goal of the project was to improve the institutional framework for land administration, which included a systematic review of existing legislation, the drafting of new land laws and regulations, and the training of staff of the National Land Agency (Badan Pertanahan Nasional or BPN, hereafter NLA) (World Bank, 1994).3 The LAP particularly focused on West-Java, as this region had the lowest coverage of titled land on the island (13%) at the time the project was initiated. Therefore, of the 1.2 million certificates planned under the LAP, 68% was to be issued within West-Java; including Bandung Municipality, with 2.5 million inhabitants Indonesia’s fourth largest city (World Bank, 1994).4

1 In some regions so-called Regional Agrarian Project Operations (Proyek Operasi Daerah Agraria or PRODA) were launched. These programs were similar to the PRONA but financed by the regional governments. In addition, the Indonesian government initiated a village-by-village titling program, the PP10 program. 2 See General Consideration, under c, Decision of the Minister of Home Affairs No. 189/1981. 3 Total costs of the project were budgeted at USD 140.1 million (World Bank, 1994). 4 Although not all eligible land parcels in Bandung have been titled under the project, still substantially more certificates have been issued than originally planned, namely 132,863 (Internal document of the Bandung Municipality Land Office).

The successor of the LAP, the Land Management and Policy Development Project, was initiated in 2004 and runs until the end of 2009. This project is even more ambitious than the LAP and has a stronger focus on land titling and institutional development than its predecessor. The goal of the project is to issue 2.5 million certificates in Java and urban centres outside Java (World Bank, 2004). Bandung’s low-income settlements: kampongs As in the other cities where the LAP has been implemented, in Bandung the project primarily targeted people residing in kampongs, Indonesia’s traditional low-income settlements. The first kampongs in Bandung were originally rural villages that were engulfed by the expanding city early 20th century. Throughout the colonial period these settlements became increasingly densely populated. When, after the struggle for independence, the West-Java countryside was confronted with warfare between the Darul Islam and the Republican Army, Bandung’s kampongs became the destination of vast flows of refugees fleeing the conflict areas. Coinciding with the exodus of the European population from Indonesia, the refugees settled in existing kampongs or created new kampongs by occupying vacant land. In later years, particularly during the New Order period (1966–1998), kampongs continued to grow as a result of labour migration, thus developing into low-income settlements with high population densities and possessing the main characteristics of informal settlements in developing countries.5 Over the years, the living conditions in kampongs have improved significantly, not least because of the government’s Kampung Improvement Program (KIP), which focussed on the improvement of infrastructure and the provision of services and later also addressed socio-economic issues. The initiators of the KIP supposed that the improvement of the kampong’s infrastructure would lead dwellers to take the initiative to improve their tenure status themselves. There is however no evidence that this has actually happened.6 Different types of tenure in kampongs In accordance with the origins and development of kampongs, three categories of land tenure are found in these settlements: formal land tenure, semi-formal tenure, and informal tenure. Dwellers with formal tenure have a property title to their land on the basis of rights acknowledged by the 1960 Basic Agrarian Law (BAL), which remains the general framework for Indonesian land law.7 At the same time, semi-formal tenure is still common in kampongs. Throughout the colonial period, because of legal dualism, residents could apply their own traditional adat law in relation to land. In kampongs this adat law soon took form of individual, inheritable adat ownership rights. Under the BAL, the position of adat law changed as part of a unification effort. It does not acknowledge adat ownership rights. Yet these rights do have a semi-formal status, as they are acknowledged as legitimate claims for formalisation (to be explained shortly). Finally, a third category, informal land tenure, is found in kampongs. Informal tenure is usually the result of dwellers squatting land on which the state holds a direct right of avail (state land) or, less often, land on which others have established private rights. Dwellers in this category therefore have no legal basis for their tenure.

5 These characteristics are: precarious sub-standard building, a lack of adequate infrastructure and public services, and an informal character in terms of land tenure and land use. 6 See for instance Taylor (1987), who in his evaluation of the Jakarta Kampong Improvement Program was not able to draw any conclusions on this matter. 7 The word ‘agrarian’ (agraria) does not imply that the law is about agriculture or agricultural land; rather, it is a reinterpreted term from colonial times that refers to land, water, and airspace (compare Art. 5 BAL).


ARTICLE IN PRESS G. Reerink, J.-L. van Gelder / Habitat International xxx (2009) 1–8

In the three different tenure categories one can see reflected the legal continuum idea discussed earlier in which legality is more a matter of degree than a strict dichotomy as portrayed by the titling approach. Semi-formal tenure takes an intermediate place between formal tenure and informal tenure. Both semi-formal and informal tenure can be formalised through titling programs such as the LAP. Semi-formal tenure can be formalised through the legal conversion of ‘old’ colonial adat ownership rights, whereas informal tenure can be formalised by granting ‘new’ rights over state land to its occupants.8 Indonesian land law does not acknowledge the concept of adverse possession - acquisition of title by peaceful occupation of land for a statutory period. Informal tenure, therefore, cannot be formalised on grounds of extended occupation of land on which others have established private rights. Such private rights may however be forfeited, for instance through neglect of the land. In this case, land passes to the state, over which it can again grant ‘new’ ownership rights (Harsono, 2005).9 In practice, mostly semi-formal tenure, not informal tenure, is formalised through titling programs like the LAP (see also Hardjono, 1999).10 According to the Head of Bandung’s Land Office, informal tenure mostly remains untitled because people have no legal basis for occupying it and thus cannot meet the evidence requirements to obtain a certificate.11 As explained above, even though the government could grant ‘new’ rights or transfer its own rights to these informal residents, there appears to be no political will to do so. Instead of granting ‘new rights’ to informal residents, the NLA has been granting such rights to public entities, like the municipality, which claims 51% of the city’s land.12 The municipality could, in turn, transfer these rights to informal residents, but this only occurs on a limited scale.13 It does grant permits for the use of the land to informal residents, thus generating significant retributions while at the same time keeping the land ‘available’ for future development projects. Identification by the state of neglected land, which results in forfeiture of rights, and redistribution of such land to low-income people appears never to occur, although much squatted land qualifies as such.14 Land titling programs: benefiting the needy? Although land titling programs continue to be implemented in (urban) Indonesia on a large scale, few evaluations have been conducted into their effects on perceived tenure security and

8 Art. 22(2), 31, 37, 41 Law No. 5/1960 in conjunction with Regulation of Department of Home Affairs No. 6/1972, Government Regulation No. 24/1997 and, Regulation of the Department of Home Affairs No. 5/1973. 9 Government Regulation No. 36/1998 in conjunction with Art. 27, 34, and 40 BAL. 10 Hardjono does however refer to one case in Semarang, where squatters of state land that was managed by the Indonesian Railway Company (PT Kereta Api Indonesia or PT KAI) were granted new rights. She suspected that political motives influenced this decision, which was taken shortly before the 1997 General Elections (Hardjono, 1999). 11 ’BPN: 30 Persen Lahan di Kota Bandung Belum Bersertifikat’, Republika, 28 November 2005. 12 On the basis of Presidential Decree No. 32/1979, informal residents hold priority rights to obtain new rights on much of this land, but they appear to be unaware of this and the state simply ignores it. 13 As far as the municipal government is willing to transfer the land, residents must actually first prove that they have a permit to use the land and have paid retribution for it. See Art. 3–7 Regional Regulation of Bandung Municipality No. 24/ 2001. 14 A good example of such land in Bandung is land over which the Indonesian Railway Company has a management right. The company manages a total of 269,900,000 Ha. in the whole of Java. Only 1140.75 Ha. or .42% is leased out to people (interview with senior manager of the Indonesian Railway Company, 19 January 2005). Most of the land that has not been leased out is simply neglected, which explains why so many people can squat it.

3

improvements to property. The existing research consists of two World Bank assigned evaluations of the LAP by Hardjono (1999) and Smeru (2002), and an independent study by Winayanti and Lang (2002).15 In relation to the effects of titling on perceived tenure security, the two World Bank assigned evaluations are rather positive. According to Hardjono (1999), participants of the LAP ‘‘almost invariably said that they feel more secure about their rights to land.’’ This perception could be related in large part to a decrease in ‘external’ threats: participants argued that ‘‘there is now no possibility of being evicted from their land against their will for development projects of any kind’’ (Hardjono, 1999; 30). According to Smeru (2002), about 70% of the participants believed that they now enjoyed greater tenure security, ‘‘as a land certificate now recognises their ownership rights’’ (Smeru, 2002; 25–26). Winayanti and Lang (2002) argue that perceived tenure security does not always have to be based on property title. The semi-formal and informal residents in the kampong in East Jakarta where they conducted their research perceived their tenure to be more secure because of a government policy of condoning non-formal tenure and providing public services, in conjunction with a high degree of community organization and the support that residents received from civil society groups. Yet, despite enjoying de facto tenure security, these residents were still hoping for legal recognition to further strengthen their tenure security. The above studies reported somewhat different outcomes regarding the effect of titling on improvements to property. Hardjono (1999), even though reporting stronger perceptions of tenure security by LAP participants, did not find a clear relation between titling and improvements to land. She explained this by stating that ‘‘the reason why [participants] had not improved their property in the past was lack of money rather than a lack of secure tenure’’ (Hardjono, 1999; 32).16 Smeru (2002) did report an overall difference in the proportion of improvements. In urban areas, however, this reported difference was marginal.17 The difference in outcomes between the studies of Hardjono (1999) and Smeru (2002) may be related to the fact that the latter study was conducted three years later, when the effects of titling on housing consolidation were more likely to be more manifest. Winayanti and Lang (2002) argued that in the kampong where they conducted research, the strengthening of de facto tenure security resulted in a substantial increase in housing improvement. Although the above studies give valuable insight into the issues under discussion, they all have methodological limitations. First, one could question whether the effects of titling would be discernable immediately after (Hardjono, 1999) or just three years after (Smeru, 2002) the LAP was implemented; particularly with regard to housing consolidation, which requires significant financial

15 The study by Hardjono was conducted in collaboration with a number of local NGOs and had a qualitative character, using methods like focus group discussions and in-depth interviews with members of communities, and was conducted in nine urban, peri-urban, and rural city quarters and villages, including a city quarter in Bandung (Hardjono, 1999). The study by the Indonesian research institute Smeru was quantitative in nature and based on a comparative survey among LAP and non-LAP participants in various villages and city quarters in 14 districts and cities, including a total of 110 participants from two city quarters in Bandung (Smeru, 2002). The study by Winayanti and Lang (2002) was based on qualitative research in the low-income kampong Penas Tanggul, in City Quarter Cipinang Besar Selatan in East Jakarta. 16 However, in one kampong in Semarang where informal residents had been granted ‘new’ rights on state land, they immediately started undertaking improvements to their property, including housing. 17 Only 3.5% more plots had been improved in areas where the LAP had been implemented than in areas where it had not been implemented and the paper did not clarify whether this observed difference was in fact statistically significant.


ARTICLE IN PRESS 4


resources. Moreover, the study by Winayanti and Lang (2002) concentrated on a single kampong, which limits the representativeness of their findings. The studies by Hardjono and Winayanti and Lang were also qualitative in nature, and lacked a statistical basis to accurately assess the effects of the titling program. Although Smeru (2002) did conduct a more representative and quantitative study, as in Hardjono’s (1999) study, no distinction was made between semi-formal and informal tenure, thus ignoring potential differences in effects of titling on these two tenure categories. The present paper overcomes these shortcomings and examines the relations between titling, perceived tenure security and housing improvement using a survey in seven kampongs of Bandung. However, first reference is made on Indonesia’s changing political-legal context in which land titling programs are being implemented.

city. As Fig. 1 shows, some of these settlements were centrally located, while others were situated more towards the outskirts of the city.19 Furthermore, each of the selected kampongs had developed incrementally over time, and housed several thousand lowincome migrants and their offspring, most of whom had been living there for various decades. All seven settlements had been targeted by the KIP or one of its successors. As a result, all had similar levels of infrastructure and most residents had access to the same types of public services. The provision of infrastructure and services was not related to tenure status.20 While all residents benefited equally from the programs, the various titling projects consisted of the allocation of property titles only and did not involve physical upgrading, so that it is possible to isolate the effects of property titles. Finally, in each of the researched kampongs the different categories of formal, semi-formal, and informal land tenure were represented.

Land titling programs and Indonesia’s broader land policy Participants and procedure While the Indonesian government continues to implement land titling programs to increase tenure security for economically vulnerable residents, it has so far maintained a body of legislation that creates room for itself and also private parties to involuntarily remove particularly this kind of residents against no or very little compensation - no matter whether holding title or not. Most of this eviction legislation was enacted in the early 1960s, before Suharto seized power in 1965 and established his New Order regime, which was marked by authoritarian rule, a ‘developmentalist’ policy of state-led industrialisation and large-scale exploitation of natural resources, and corruption. During the New Order, new regulations and government policies enabled the government and wellconnected businessmen to take whatever land they needed. It must been noted, however, that Indonesia’s political-legal context has changed significantly in recent years. President Suharto’s fall in 1998 brought an end to the New Order regime and much of its policies. An ambitious reform program, which included one of the world’s most radical decentralisation initiatives, resulted in Post-New Order Indonesia developing into a state in which signs of an emerging rule of law are easy to identify - although corruption remains a significant problem.18 The above reforms have not unequivocally improved the de facto tenure security of low-income kampong dwellers in Bandung. Fiscal aspects of decentralisation have contributed to the municipal government stimulating commercial land development, which results in more residents being confronted with parties who want their land. At the same time they dare to resist involuntary removal. While generally failing to prevent this, these kampong dwellers do succeed in negotiating higher compensation – even those without title.

The survey sample consisted of a total 340 households across the three mentioned tenure categories: formal tenure, as indicated by titled households (100 households), semi-formal tenure of households with adat ownership rights (95 households), and informal tenure of households squatting state land or, less often, private land (145 households).21 The surveys were distributed approximately equally across the seven selected kampongs, so that about 50 lowincome households were surveyed in each kampong.22 Data collection took place during September, October, and November 2005. See Table 1 for further details on the households surveyed. Materials Survey items were developed on the basis of the previously discussed literature, participant observation in two kampongs, and interviews with NGO workers, government officials, and researchers in Bandung. To check the accuracy of information provided by participants, the survey contained several questions related to land documentation (e.g. the year such documents were issued, the identities of issuing authorities, the presence of witnesses, and the period of document validity), and questions about the length of occupation and formal status of the land. Participants were also asked to show their land documentation to the surveyors, to verify whether it corresponded with the given answers. The surveys of participants who gave inconsistent answers were excluded from further analyses. The specific questions asked during surveys to investigate perceptions of tenure security and to measure housing consolidation will be presented and discussed in the next section.

Method Location To obtain representative data on perceived tenure security and housing consolidation in Bandung, a survey was conducted in seven kampongs of roughly the same size located in different parts of the

18 Indonesia’s amended 1945 Constitution now recognises the right to adequate housing, of which tenure security forms an important element. Furthermore, the country has become party to various international human rights treaties that recognise this right, including the International Covenant on Economic, Social and Cultural Rights. It goes without saying that these steps have created an inconsistency in relation to the previous discussed legislation that allows the state and private parties to involuntarily remove residents against no or very little compensation. To date, there are no indications that the government is planning to address this inconsistency.

19 The kampongs were located in the following City Quarters (Kelurahan): Babakan Surabaya, Cibangkong, Ciroyom, Kebon Lega, Kebon Pisang, Lebak Gede, and Taman Sari. 20 Officially KIP was only supposed to be implemented in ‘legal’ settlements (i.e. in conformity with zoning regulations), but in practice the program also entered in non-conforming settlements. 21 Participants with formal tenure also included residents who had acquired their land title themselves, instead of through a land titling program. As all survey results were identical for both these categories of participants, they were combined into a single one tenure category. 22 Within each kampong, to obtain representative data several, Administrative Neighbourhoods (Rukun Warga or RW) were randomly selected, and several administrative Housing Blocks (Rukun Tetangga or RT) were then randomly selected within each Administrative Neighbourhood. Within each Housing Block, survey participants were selected to ensure similar representation across tenure categories. This was achieved with the assistance of Heads of Neighbourhood or Heads of Housing Blocks, who were able to advise into which tenure category residents fell.


ARTICLE IN PRESS 5

Fig. 1. Map of Bandung with survey locations.



ARTICLE IN PRESS 6


Table 1 Gender, age, household income and housing consolidation (N ¼ 340). Variables

Table 2 Is there a possibility of involuntary removal within the next 5 years (N ¼ 340).

Legal tenure category Formal tenure (n ¼ 100)

Gender of respondent Male/Female (%) 61/39 Age of respondent M (SD) 54 (13.814) Household income ( 1000)c M (SD) 1.730 (1.461)a Housing consolidationd M (SD) 8.78 (.52)a

Semi-formal tenure (n ¼ 95)

Informal tenure (n ¼ 145)

66/34

71/29

53 (12.676)

49 (11.957) a/b

1.401 (1.211)

1.256 (1.255)b

8.56 (.80)a/b

8.32 (1.00)b

a,b

Means in the same row that do not share a superscript letter differ at p < .01. Household income in Indonesian Rupiahs (Rp. 10,000 z USD 1). d Rated as a score from 3 to 9, with 3 representing the lowest level of consolidation and 9 representing the highest; see Results section for details. c

Results Perceived tenure security We first consider the differences between the three tenure categories with respect to two indicators of tenure security. As a first indicator, participants were asked whether they believed there was a possibility of involuntary removal within the next five years. As Table 2 shows, both formal and semi-formal residents perceived a possibility of involuntary removal, but much less often than did informal residents.23 It should however be noted that more than 76% of the informal residents also perceived no possibility of involuntary removal within the next five years, despite their informality. We asked those residents who thought involuntary removal was possible to specify what they believed might cause it. Land acquisition in the public interest was most often mentioned by residents in all tenure categories. In addition, land acquisition for commercial purposes was mentioned by several semi-formal residents, and eviction for occupying land without permission from the titleholder was mentioned by several informal residents. These findings demonstrate that involuntary removal is perceived by residents from all tenure categories to be driven primarily by the state. The second indicator of tenure security refers to the perceived legitimacy of tenure - that is, whether residents think that the authorities agree with them residing on the land they occupy. Participants’ responses to this question are presented in Table 3, and indicate that a higher percentage of both formal and semi-formal residents believed that the government agreed with their residence than did informal residents (90% for formal and semi-formal residents versus 70% for informal residents).24 The results indicate that residents with a legally stronger tenure category have stronger convictions that the government agrees with their residence, while residents with informal tenure are less confident of agreement, and more likely reply ‘maybe agrees’. Yet it is noteworthy that a large majority of informal residents still thought that the government agreed with their residence, and that again the difference between forma and semi-formal households is negligible. Participants who reported that the government agreed with their residence were asked why they believed this to be the case. As Table 4 shows, the possession of land related documentation was the primary reason given by most formal residents (91.2%).

Possibility No possibility

Formal tenure (n ¼ 100)



5.3% 94.7%

6.9% 93.1%

23.6% 76.4%

Table 3 Does the government agree that you reside on this land? (N ¼ 340). Government agreement

Formal tenure (n ¼ 100)



Yes Maybe No

88.9% 7.1% 4.0%

91.6% 5.3% 3.2%

70.1% 26.4% 3.5%

Surprisingly, the possession of documentation was also mentioned as a primary reason given by a large percentage of semi-formal (66.7%) and informal residents (47.2%). For semi-formal and informal residents, the length of occupation (which as mentioned above is an important element of de facto tenure security) was another important reason to believe that the government agreed with their residence (around one third of the respondents). Servicing - another factor often assumed to contribute to de facto security - was much less frequently mentioned (less than 5% of respondents across tenure categories). Housing consolidation To investigate whether investment in housing (as indicated by the level of housing consolidation) was associated with tenure status, we tested for differences in the means of housing consolidation scores between tenure categories. Housing consolidation was measured on the basis of a composite score which represented the quality of the three defining elements of a dwelling; namely the floor, the walls, and roof. For each element a three-point index of consolidation was used, with a score of 1 indicating low consolidation, and 3 representing high. The scores were combined into a composite score that could range from 3 (indicating low consolidation) to 9 (indicating high consolidation). As Table 1 shows, there were differences in levels of housing consolidation between the different tenure categories.25 Residents in formal tenure situations lived in more consolidated housing than informal residents.26 There were, however, no significant differences in degree of housing consolidation between semi-formal and formal residents. The difference between semi-formal and informal residents was marginally significant.27 As shown in Table 1, there were also differences in household income between the different tenure categories. Residents with formal tenure had significantly higher incomes than residents with informal tenure, but again no significant differences in income were found between formal and semi-formal residents, nor between semi-formal and informal residents.28 As a final step in the analysis we employed a regression analysis, using household income, perceived possibility of involuntary

25

F (2, 337) ¼ 9.420, p < .001). Tukey’s HSD, p < .001. p ¼ .071. 28 An analysis of variance (ANOVA) test demonstrated that these differences were significant (F(2,337) ¼ 3.939, p < .05). To examine potential differences between particular categories, post-hoc tests using Tukey’s HSD were employed. The results revealed significant differences in income only between the formal and informal tenure categories (p < .05). 26

23

Results of the Chi-square test indicate that the difference in perceived security between tenure categories is significant (c2 ¼ 19.908, df ¼ 2 p < .001). 24 Results from the Chi-square test indicate that, with respect to perceptions about government agreement (that government agreed with residence, maybe agreed, or did not agree), relative frequencies of the three answers were significantly different between tenure categories (c2 ¼ 27.142, df ¼ 4, p < .001).

27


ARTICLE IN PRESS G. Reerink, J.-L. van Gelder / Habitat International xxx (2009) 1–8 Table 4 Why do you believe the government agrees with you residing on this land? (N ¼ 287). Reason

Formal tenure Semi-formal Informal (n ¼ 91) tenure (n ¼ 90) tenure (n ¼ 106)

Because I have land related 91.2% documentation Because I have been occupying 4.4% this land for a long time Because the government 3.3% provides public services Other 1.1%

66.7%

47.2%

31.1%

36.8%

2.2%

4.7%

0%

11.3%

Table 5 Regression of household income, perceived possibility of involuntary removal and tenure category on housing consolidation (N ¼ 328). Predictor variables

Regression analysis B

SE B

b

Household income Perceived possibility involuntary removal Tenure category Formal vs. semi-formal Formal vs. informal

.07 .32

.00 .12

.15** .14*

.18 .37

.12 .11

.10 .22**

Note. R ¼ .30, R2 ¼ .09, SE (.82); *p < .05, **p < .01.

removal, and legal tenure category as predictors of housing consolidation, to be able to determine the unique contribution of each predictor to levels of housing consolidation (Table 5).29 The results show that both household income and perceived possibility of involuntary removal were significant independent predictors of housing consolidation. For tenure category, the regression analysis confirms the findings of the ANOVA: dwellers with formal tenure live in significantly more consolidated dwellings than do informal dwellers, but again no difference was found between formal and semi-formal dwellers. Therefore, formal residents can be differentiated from informal residents with respect to their level of investment in housing, but there were no such differences between formal and semi-formal residents. Finally, the effect for legal tenure category on housing improvement cannot be explained by differences in household income between the different tenure categories, and perceived possibility of involuntary removal is also an independent indicator of housing consolidation that is not explained by legal category. Conclusion As part of the broader debate on the effects of land titling, this paper assessed its effects on perceived tenure security and housing consolidation in kampongs in the city of Bandung, Indonesia. Titling programs generally operate under the assumption that residents with property title enjoy greater tenure security than residents living in ‘extra-legal’ tenure situations, and as a result are also more willing to invest in their housing. While titling has been widely applied worldwide, its underlying assumptions have been criticised for being oversimplified, and its assumed benefits for being exaggerated. Our findings support the assumption that titling contributes to perceived tenure security for residents in Bandung’s kampongs. The perceived tenure security of residents with a property title was higher than that of informal residents. However, no such differences were found between residents with title and semi-formal

29

Due to its categorical nature, tenure status was dummy-coded.

7

residents. These findings, in combination with the observations that semi-formal (and informal) dwellers often cited their land related documentation and length of residence as reasons for the legitimacy of their tenure, support the idea that perceived tenure security is enhanced not only by titling, but also by increasing de facto tenure security. Moreover, they substantiate the argument that tenure should not be categorised in dichotomous terms, and that directly equating formal tenure with security and informality with insecurity is too simplistic. We also found that titling did contribute to housing consolidation - but only in a marginal way. Kampong dwellers with formal tenure lived in somewhat more consolidated dwellings than those with informal tenure. But again, no significant differences were found between formal and semi-formal residents. From the above we conclude that titling programs strengthen the perceptions of tenure security and contribute to housing consolidation in Bandungs’ kampongs, but only in as far as their efforts concentrate on the formalisation of informal tenure. This implies that titling programs in Bandung and elsewhere in urban Indonesia have so far been targeting the wrong group; as the people who may benefit most from titling programs, residents with informal tenure, rarely find themselves among their beneficiaries. As Bromley (2008; 22) argued, by issuing titles, governments appear to legitimise the presence of individuals and their dwellings and show signs of being prepared to take them seriously. If the Indonesian government and donors are indeed seriously committed to increasing the perceived tenure security of economically vulnerable people and enhancing housing consolidation in kampongs, land titling programs should be directed at the formalisation of informal tenure and be combined with a fundamental review of Indonesia’s outdated eviction legislation. Acknowledgments We greatly acknowledge the help of the following people who have assisted us with the survey research: Dede Tresna Wiyanto, Ade Sudrajat, Yadi Suryadi, Ivan Rahadian, Deni Kurniawan, and Helmi Suryanegara. During the fieldwork, the first author was a visiting researcher at the Faculty of Law of the Parahyangan Catholic University in Bandung. He is much indebted to Dean Ismadi Bekti, fellow researcher Tristam Moeliono, and other staff for their hospitality as well as the excellent research environment they provided. We are also grateful to the Society for the Advancement of Research in the Tropics (Treub-Maatschappij), the Netherlands Organization for Scientific Research (NWO), the Adat Law Foundation (Adatrecht Stichting) and the Leiden University Fund (LUF), which gave financial support for the fieldwork in Bandung. References Broegaard, R. J. (2005). Land tenure insecurity and inequality in Nicaragua. Development and Change, 36, 845–864. Bromley, D. W. (2008). Formalising property relations in the developing world: the wrong prescription for the wrong malady. Land Use Policy, 26, 20–27. Calderoń, J. A. (2004). The formalisation of property in Peru 2001–2002: the case of Lima. Habitat International, 28, 289–300. De Soto, H. (2000). The mystery of capital: why capitalism triumphs in the West and fails everywhere else, New York. Durand-Lasserve, A. (2006). Informal settlements and the Millenium Development Goals: global policy debates on property ownership and security of tenure. Global Urban Development, 2(1), 1–15. Durand-Lasserve, A., & Royston, L. (2002). International trends and country contexts - from tenure regularization to tenure security. In A. Durand-Lasserve & L. Royston (Eds.), Holding their ground, secure land tenure for the urban poor in developing countries (pp. 1–34). London. Durand-Lasserve, A., & Selod, H. The formalisation of urban land tenure in developing countries. Paper prepared for the World Bank’s fourth Urban Research Symposium, Washington, DC, May 14–16, 2007.


ARTICLE IN PRESS 8


Fernandes, E., & Varley, A. (1998). Law, the city and citizenship in developing countries: an introduction. In E. Fernandes, & A. Varley (Eds.), Illegal cities: law and urban change in developing countries. London. Gilbert, A. G. (1994). The Latin American city, London. Gilbert, A. G. (2002). On the mystery of capital and the myths of Hernando de Soto: what difference does legal title make? International Development Planning Review, 16, 1–19. Hardjono, J. (1999). A social assessment of the land certificate program, the Indonesian Land Administration Project, Unpublished report prepared for the World Bank, Jakarta. Harsono, B. (2005). Hukum agraria Indonesia, sejarah pembentukan Undang-Undang Pokok Agraria, isi dan pelaksanaannya, Jakarta. Kundu, A. (2004). Provision of tenurial security for the urban poor in Delhi: recent trends and future perspectives. Habitat International, 28, 259–274. Pamuk, A. (2000). Informal institutional arrangements in credit, land markets and infrastructure delivery in Trinidad. International Journal of Urban and Regional Research, 24, 379–396. Payne, G. (1997). Urban land tenure and property rights in developing countries: a review, London. Payne, G. (2002). Conclusion: the way ahead, In G. Payne (Ed.), Land, rights and innovation, improving tenure security for the urban poor, London, 300–331. Payne, G. (2001). Urban land tenure policy options: titles or rights? Habitat International, 25, 415–429. Payne, G., Durand-Lasserve, A., & Rakodi, C. Social and economic impacts of land titling programmes in urban and peri-urban areas: a review of the literature. Paper prepared for the World Bank’s fourth Urban Research Symposium, Washington, DC, May 14–16, 2007. Razzaz, O. M. (1993). Examining property rights and investment in informal settlements: the case of Jordan. Land Economics, 69(4), 341–355.

Sjaastad, E., & Cousins, B. (2008). Formalisation of land rights in the South: an overview. Land Use Policy, 26, 1–9. Smeru (2002). An impact evaluation of systematic land titling under the Land Administration Project (LAP), Jakarta. Taylor, J. L. (1987). Evaluation of the Jakarta Kampung improvement program, In R. J. Skinner, J. L. Taylor, & E. A. Wegelin (Eds.), Shelter upgrading for the urban poor, evaluation of Third World experiences (pp. 39–67). Manila. Turner, J.F.C. (1976). Housing by people: towards autonomy in building environments. London. Van Gelder, J. L. (2007). Feeling and thinking: quantifying the relationship between perceived tenure security and housing improvement in an informal neighbourhood in Buenos Aires. Habitat International, 31, 219–231. Van Gelder, J. L. (2009). Legal tenure security, perceived tenure security and housing improvement in Buenos Aires: an attempt towards integration. International Journal of Urban and Regional Research, 33, 126–146. Varley, A. (1987). The relation between tenure legalization and housing improvements: evidence from Mexico City. Development and Change, 18, 463–481. Varley, A. (2002). Private or public: debating the meaning of tenure legalization. International Journal of Urban and Regional Research, 26(3), 449–461. Von Benda-Beckmann, F. (2003). Mysteries of capital or mystification of legal property? European Journal of Anthropology, 41, 187–191. Winayanti, L., & Lang, H. C. (2002). Provision of urban services in an informal settlement: a case study of Kampung Penas Tanggul, Jakarta. Habitat International, 28, 41–65. World Bank. (1994). Staff appraisal report Indonesia Land Administration Project, Washington, DC. World Bank. (2003). Land policies for growth and poverty reduction, a World Bank policy research report, Washington, DC. World Bank. (2004). Project appraisal document Land Management and Policy Development Project, Washington, DC.
