1. INTRODUCTION
This paper uses the development of land law in Cambodia to identify the nature, process, and risk inherent in foreign-led legal construction in countries without the robust social, political, and institutional structures that some would consider prerequisites for an effective legal system. We chose land law because it is at the intellectual and institutional centre of efforts to build legal systems in poor countries. Security in land tenure is generally assumed to be necessary for optimum levels of agricultural productivity, and the provision of formal legal title to land has been proposed as the best way to provide such security.Footnote 2 We chose Cambodia because as an agricultural country whose future depends on the success of its rural sector, it is representative of many developing countries facing pressures to reform their land law. It has also been the recipient of sustained legal assistance from a variety of donors over the last three decades. Successive waves of foreign advisors have not only designed the relevant statutes, but have also funded and overseen the building of the bureaucracies charged with their implementation. The breadth, depth, and duration of foreign presence combine to make Cambodia an unusually apt and accessible site to investigate both the utility of foreign intervention in law reform and the interaction between legal rules and the public bureaucracies designed to enforce them.
This paper builds on the literature assessing the relationship between property rights and economic development. It is generally understood that secure property rights contribute to growth by, inter alia, incentivizing investment,Footnote 3 facilitating asset transfer,Footnote 4 and enabling entry into the commercial credit markets.Footnote 5 There is less agreement, however, on the best method to achieve property security. Hernando de Soto, famously, promotes centralized, government-led codification of ownership in the form of formal land titles, often drawing on foreign expertise and technology.Footnote 6 His ideas have had enormous influence in the developing world, including, as we shall see, on Cambodia’s Land Law of 2001. Others argue for more incremental, local, and informal methods and are sceptical of universal models and “best practices.”Footnote 7 This approach can be seen in the relevant provisions of Cambodia’s Civil Code of 2007.
Empirical evidence can be brought to bear on both sides of the debate, and it is important to note that the two approaches are not mutually exclusive. Indeed, although this paper paints a sombre portrait of foreign intervention, we do not argue conclusively for or against titling programmes as exemplified by the Land Law, and it would be ridiculous to think that a country in Cambodia’s situation at the turn of the twenty-first century could or should turn its back on foreign experience or guidance. Instead we offer this case-study as a cautionary tale. Although we assume that legal borrowing is both inevitable and desirable, not only for countries in Cambodia's situation but for any country trying to improve its legal system, we argue that legal transplants present just as many dangers as any other form of social engineering. No matter how successful, it is unlikely that any transplant has ever functioned in its new context as it did in the old. They will always be distorted by the political and bureaucratic power struggles of the recipient society, which will be heightened in instances like Cambodia where the foreign sponsors themselves are deeply divided. When the destination society lacks strong bureaucratic institutions and the political structure is at best immature, the dangers multiply. Well-intentioned and well-designed policies will flounder on the shoals of bureaucratic incompetence and corruption, and the legitimate interests of large numbers of people will be ignored. This paper’s contribution, therefore, lies not in providing closure to this debate, but in providing an account of how context, including the interaction of domestic and international bureaucratic and political rivalries, can and will affect the outcomes of attempts to create a new legal order.
The paper proceeds as follows. Part 2 summarizes Cambodian land use practice and legislation up to 2012. It focuses on conflict between the Land Law of 2001 and the Civil Code of 2007. The former was drafted by a World Bank-sponsored team and introduced an Australian-based system that emphasizes clarity and simplicity in land titling by recognizing ownership exclusively on the basis of formal registration with a centralized cadastral agency. The Japanese-drafted Civil Code, on the other hand, treats registration as establishing only a rebuttable presumption of ownership. These contrasting approaches not only reflect the national experiences of their drafters but also represent two opposing views of law and its role in social and economic development: Should legal rights—in this instance ownership of land—be made simple, clear, and universal so that assets can be easily exchanged in Coasian bargaining, an approach that some have characterized as “bright-line fever?”Footnote 8 Or should legal rights reflect and reinforce established local norms so that they integrate more readily into existing social practice? Part 3 attempts to shed light on these alternative approaches by examining the current state of Cambodian land rights and land policy implementation. Throughout, the paper notes the ongoing tensions among Cambodia’s various foreign patrons and the domestic ministries that have aligned with one side or the other. Part 4 concludes with a tentative assessment of foreign involvement but without any pretence of offering failsafe prescriptions or best practices for legal reform in poor countries.Footnote 9
2. THE EVOLUTION OF CAMBODIA’S LAND LAW
Until French colonization in the late nineteenth century, land in Cambodia was the property of the king, but individual subjects had rights to possess and use land and to pass it to their heirs as long as they continued to cultivate it.Footnote 10 Individual rights to land that had been cleared and used were recognized by royal proclamation when Cambodia became a French protectorate,Footnote 11 but the concept of fully private property in land in the European sense was first introduced by the Civil Code of 25 February 1920,Footnote 12 which enabled occupants to submit a request to a designated official to recognize their private rights. By 1930 most rice-growing land was registered as private property,Footnote 13 but lack of bureaucratic capacity meant that only 10% of landowners received ownership titles, with the rest receiving certificates of possession.Footnote 14 This system of land administration remained largely unchanged through independence in 1953 and up to the Khmer Rouge takeover in 1975, when all laws became irrelevant.
The Khmer Rouge abolished all private property in land and destroyed all records of ownership. The state owned all land and allocated it for use by “solidarity groups” in a massive socialist experiment. The experiment failed. As many as two million people died during the Khmer Rouge’s five years in power, and annual rice production declined to less than one million tons (as compared to more than seven million tons in 2010). These material costs were exacerbated by the elimination of almost all educated people, so that by the time that the Vietnamese invaded in 1979 and established the People’s Republic of Kampuchea (PRK) with Hun Sen at its head, there remained fewer than ten lawyers and 50 doctors and virtually no one with the legal or technical capacity to administer a cadastral system. Whether for that reason or ideological ones, the system of land management did not change significantly. Land continued to be owned by the state and used by solidarity groups. Families were allowed small plots, but there was no effort to restore private ownership, much less pre-1975 ownership rights.
2.1 The Land Laws of 1992 and 2001
With the departure of the Vietnamese in the late 1980s and the creation of the United Nations Transitional Authority in 1991, the Hun Sen regime began the establishment of a legal framework appropriate for a market-oriented democratic state. With elections scheduled for 1993, the immediate goal was the minimum necessary to establish a sense of order.Footnote 15 The Land Law of 1992 was the product. Because of the urgency in drafting, the government promulgated the Law without much discussion, and many of the provisions were simply copied from the 1920 Civil Code.
A more deliberative approach was possible for the drafting of the 2001 Land Law. The goals for the legislation were the integration of Cambodia into the world economy and the provision of greater tenure security to average Cambodians.Footnote 16 To accomplish these goals, simplicity and transparency were considered of paramount importance. In keeping with this priority, the law not only disavowed any attempt to re-establish pre-1975 property rights, but also allowed people in lawful possession of land as of 2001 to apply for ownership after five years of peaceful, continuous, unambiguous, and open possession in good faith. What the law emphatically did not allow was for post-2001 possession or other forms of use to ripen into ownership.
Instead, the drafters looked to Australia’s Torrens system and created a regime where title was to be determined exclusively by registration in a cadastral registry to be established under the Ministry of Land Management, Urban Planning, and Construction (MLMUPC or Land Ministry). Article 239 states that cadastral records “have legal value and precise effect,” which has been interpreted to mean that the registry prevails even when registration was mistaken or fraudulent. Consistent with this interpretation, Article 226 states that “ownership of immovable property shall be guaranteed by the State,” which has been interpreted to mean that rightful owners whose interests were not accurately represented by the registry would receive governmental compensation but would not regain ownership of the land. Most conclusively, Article 65 states:
The transfer of ownership can be enforceable as against third parties only if the contract of sale of immovable property is made in writing in the authentic form drawn up by the competent authority and registered with the Cadastral Registry Unit. The contract of sale itself is not [sufficient] for the transfer of the ownership of the subject matter.
This exclusive reliance on registration presupposes accurate and universal cadastral mapping and hopes to achieve the higher land values associated with registration systems.Footnote 17 To achieve this level of precision and thereby “reduce poverty, promote social stability, and stimulate economic development,”Footnote 18 the World Bank created the Land Management Administration Project (LMAP) to supply the technical expertise necessary for the quick, clear, and conclusive specification of ownership, commune by commune, throughout all of Cambodia.
In part in recognition of the impact of rejecting post-2001 prescriptive acquisition of land (adverse possession in common-law systems), the Law provides for grants of land to poor Cambodians under a Social Land Concession (SLC). SLCs were seen as a more appropriate and orderly way to provide land than recognizing the informal occupation of land that characterizes land settlement in most developing countries, including Cambodia. Under the Law, SLC recipients who have often been displaced from pre-existing settlements are to receive government land. If they successfully comply with the provisions of the concession, they can then apply to the MLMUPC for ownership. In this way, the state would eliminate landlessness among poor Cambodians.Footnote 19 At the other end of the socioeconomic spectrum, the Law instituted a system of Economic Land Concessions (ELC) for agricultural or agro-industrial purposes. ELCs grant concessionaires all the rights of ownership save alienation, but are limited in size to 10,000 hectares and in duration to 99 years. Both types of concession are new constructs in the 2001 Land Law and find no mention in the 1992 Land Law or the 1920 Civil Code.
While the 2001 Land Law was formally drafted by the MLMUPC, the origin was not Cambodian. One key (French) participant described the process as follows: “Every draft was discussed by local institutions and the Cambodians tried to make it theirs but the first draft always came from the international community.” Tellingly, the law’s official domestic sponsor admitted that he “didn’t understand the law” and that it was not “our law” but “the law of NGOs.” A longtime member of the non-governmental organization (NGO) community, on the other hand, emphasized donor influence:
It depends how much noise the NGOs make about a particular issue and whether they are aware of the process before it is too late. Sometimes the donors require consultation with the community in promulgating the laws but there is still a lot of tokenism. The government just doesn’t have the political will to get input.
The precise lines of influence and causation are unimportant for our purposes. What is important is that the law is the epitome of top-down social engineering with the added dimension that it was based on foreign models and designed by foreign experts with reference to global best practices, but with virtually no knowledge of Cambodian society. It does not purport to reflect or connect to existing social practice beyond its recognition of pre-2001 possession. Unlike much such social engineering, however, it is directed not at increasing government power but the opposite: its aim is to minimize the governmental and judicial roles by creating a regime that will facilitate Coasian bargaining by private market actors. Unitary, simple, and transparent property rights would also enable the direct foreign investment that the drafters assumed would be difficult with a less centralized, less universal system.
2.2 The 2007 Civil Code
Simultaneous with the drafting of the Land Law, a parallel process was underway to draft a new Civil Code. Driven by both domestic and external pressure, the Cambodian Ministry of Justice (MoJ) commissioned the Japanese International Cooperation Agency (JICA) in 1999 to co-ordinate drafting a code. The goal was to unify the various statutes dealing with civil law including the Land Laws of 1992 and 2001, to be consistent with the Constitution of 1993, and to facilitate Cambodia’s entry into the World Trade Organization (WTO). JICA’s efforts bore fruit with the December 2007 enactment of the Civil Code, but the need for transitional provisions, exacerbated by bureaucratic competition between JICA and the MoJ on one side and the development banks and the MLMUPC on the other, meant that the Code did not go into effect until 2012.
There are many areas of uncertainty in the relationship of the Civil Code with pre-existing statutes like the Land Law, but we deal here only with the conflict between the two laws’ treatment of land registration. As discussed above, the 2001 Land Law creates a Torrens-style land registration system in which proof of registration constitutes ownership. In contrast, Article 137(1) of the Civil Code provides only that “where a right is registered […], it is presumed that such right belongs to the person to whom it is registered,” drawing into question the “legal value and precise effect” language of Article 239 of the Land Law. Equally important, Articles 162(1)−(2) allow for prescriptive acquisition of ownership of private property after 20 years of peaceful and open possession (reduced to ten years if in good faith and without negligence or fault) without reference to the pre- or post-2001 commencement of the possession.Footnote 20 Thus anyone relying only on the registry could face a dispute with the prescriptive claimants.Footnote 21 The Code, therefore, promises a means of determining ownership that is vastly more responsive to the informal facts on the ground, even if that responsiveness sacrifices the bright-line precision assumed to be necessary for market transactions.Footnote 22 The Code’s approach draws on the view that Torrens-style registration systems are only preferable when the reduction in the expected cost of forfeiture balances the higher cost of initial registration,Footnote 23 a balance that was unlikely to be met given the difficulties implementing the titling system in Cambodia.
As with the Land Law, it is worth considering the institutional origin of the Code’s rejection of the Torrens-style registration system. Japan’s own law is closer to the American deed recordation system than the Australian system, and JICA experts were worried that the land registry would quickly be corrupted by local officials or become out of date as average Cambodians failed to register subsequent transfers, as has been commonplace in titling programmes throughout the developing world.Footnote 24 JICA personnel also felt that it was inequitable to favour the bona fide third-party purchaser over the innocent original owner in the case of fraud.Footnote 25 Nor did they give credence to the Article 226 guarantee of state compensation to owners dispossessed by registry inaccuracies. Fearing low state funds and a lack of political accountability, a state guarantee was not deemed realistic.
2.3 Reconciling the 2001 Land Law and the 2007 Civil Code
To appreciate the dynamics of resolving the doctrinal conflict between the Land Law and Civil Code, one must step back and consider the political and institutional context in the years leading up to the passage of the Code. Land was a central concern from the early years of the Hun Sen regime, and in 1992 a consortium of EU governments, the UN, and other international agencies hired Finnmap, a Finnish land management consultancy, to complete a four-year aerial mapping project.Footnote 26 Finnmap next undertook a pilot land registration project in 1997, and was still engaged in related projects as of 2011.Footnote 27 The German development agency GiZ entered the land registration field in 1995, collaborating with the MLMUPC and Finnmap. After the promulgation of the Land Law, the World Bank entered the scene in 2002 with a $33.9 million investment in LMAP to map and register land.
While there was some conflict between the early and late arrivals to the mapping enterprise,Footnote 28 the MLMUPC and its various technical advisers quickly united in 2003 to oppose the entry of JICA and the Ministry of Justice and any application of the nascent Civil Code to the land policy arena. They argued that the Civil Code’s recordation system would create uncertainty, which would in turn require expensive title insurance available only to the rich, disproportionately harm poor Cambodians, and undermine the creation of the efficient land market that was the goal of the reforms.
The conflict was not solely ideological; money and power were also at stake. One advocate described the situation as follows: “Each ministry is like a separate fortified island. They don’t talk to each other and don’t work well together even in areas where their responsibilities overlap.” More concretely, in a weak bureaucracy like Cambodia’s, bribes and informal payments are commonplace. “Each signature is a tip, so if your ministry is required to sign a particular document, then you’ve just increased the budget of your ministry.”Footnote 29 Adopting the Code approach, therefore, would shift power and resources from the MLMUPC to the MoJ, and give the courts the final word on whether the presumption of authenticity granted registered title was ultimately maintained or not. The Land Ministry had a lot to lose.
After prolonged discussions over the next seven years, including some meetings in Washington, DC, among the foreign sponsors and with little Cambodian involvement, the Code was enacted. Legislative passage did not, however, conclude the matter, as several transitional issues and implementing provisions remained to be reconciled. A compromise was finally reached that left the 2001 registration system in place, but the conclusiveness of the registry was substantially weakened with various forms of possession and use becoming relevant in the determination of ownership. The MLMUPC will continue to control the registration and certification process but the courts will have the final say as to their authenticity.Footnote 30
3. The Implementation Process
The 2001 Land Law, the 2007 Civil Code, and the 2011 Law on Application of the Civil Code described above may all have been largely the product of foreign intervention, but the Cambodian government has not been totally passive. There has been no shortage of formal policy declarations, statutes, regulations, and formal internal circulars on bureaucratic practice,Footnote 31 and discussions with people involved in their drafting make clear that the process has changed over the last 10 to 15 years. Initially foreign advisors completed the first draft and only later shared it with the Ministry; today the process has been reversed, as described by one government official:
The first step is to obtain input. Typically, the ministry begins with the widest participation including local authorities, commune chiefs and district chiefs. Sometimes the public at large is invited to attend a meeting and to speak on the issue. With this information the ministry writes a first draft of the document reflecting the concerns and objectives of the various stakeholders. After a draft is established then the ministry does more narrow consultation with some of the larger national and international NGOs at higher level meetings. Next, the donor community (e.g. ADB, CIDA, JICA, WB, DANIDA, GiZ) would likely be invited to comment and to ensure that the legal text is of a high quality. These last two steps might take place in working group sessions or else through e-consultation. Finally, the draft is sent to the Council of Ministries for discussion and revisions before it is sent to the National Assembly.Footnote 32
While undoubtedly aspirational,Footnote 33 his description describes a real increase in the bureaucratic and legislative capacity of the Land Ministry, and implies a higher level of commitment to the resulting law and policies than was exhibited by the earlier externally driven process.
This is not to say that more recent legislation lacks foreign influence. Ministry staff routinely conduct literature reviews of other countries’ legal frameworks to get inspiration for their own reforms. “We are just looking for places where the situation seems to be working well and learn from that.”Footnote 34 For example, the 2010 Law on Providing Foreigners with Private Units of Co-Owned Buildings drew on Thai and Singaporean statutes to limit foreign ownership to 70% of a co-owned building above the ground floor. MLMUPC staff also added provisions limiting foreign ownership within 20 km of the border in response to government fear that relatively cheap land in Cambodia would lead to Thai and Vietnamese enclaves near the borders. This sort of adaptation to local circumstances demonstrates the Ministry’s increased confidence in policy-making and recognition of the nuanced considerations of the Cambodian context.
3.1 Registration
While technical capacity has improved and the legal framework is largely in place, the registration critical to the reconciliation of the doctrinal conflicts identified in Part 2 has lagged, although not for lack of effort. As this section explores, the land registration process reflects well-intentioned but problematic policies that have served to weaken tenure security for vulnerable parts of the population.
LMAP instituted a dual system of land registration procedures—systematic and sporadic. The former is conducted by LMAP teams in pre-selected areas and aims to map and provide titles to an entire community. The latter responds to individual applications and targets only the applicant’s land. Both are the responsibility of the Cadastral Administration, as is the creation and maintenance of the Land Registry.Footnote 35 As of 2010 these processes had reached Phnom Penh and 15 selected provinces, with registration completed in slightly less than half of those communes.Footnote 36 The systematic titling process has collected data on 2,053,062 parcels, 80% of which are rural, and has issued 1,500,493 title certificates. An additional 607,784 titles have been distributed through the sporadic process, bringing the total to 2,108,277. The process has yielded over 10 billion Riel [$2,636,760] in cadastral fees, meaning that titling has become self-financing since the 2009 withdrawal of World Bank funding.
Unfortunately these aggregate statistics mask considerable disparities within title registration. First, LMAP’s strategy was to begin systematic titling in areas that were neither “likely to be disputed” nor of “unclear status.” The rationale was to focus on areas where LMAP could be most successful, at once building administrative capacity and gaining legitimacy for the programme through early successes.Footnote 37 The result was that households and communities that lie in the path of planned developments or concessions, or whose land is desired by well-connected individuals or companies, have been excluded from the process.Footnote 38 Since neither “likely to be disputed” nor “unclear” was defined,Footnote 39 local authorities had essentially unfettered power to remove such land from the cadastral process. Since sporadic registration was considered too costly for most poor households to manage,Footnote 40 areas left out of systematic land registration remained in a state of tenure insecurity. This situation was exacerbated by the Land Law’s restriction on forms of evidence of ownership granted by prior regimes, thus exposing these communities to accusations of being illegal “anarchic squatters.”Footnote 41 Therefore, while the owners of the one and a half million parcels that have received formal title through the systematic registration have undoubtedly benefited, households left out of this initiative may be much worse off.Footnote 42
A second issue arises from the common belief that holding the title certificate, as opposed to registration with the Cadastral Administration, constitutes ownership, so that even in areas that have been successfully titled, people transfer land relying on the sale contract and delivery of certificate alone. The pattern is not new or uncommon, as the Japanese argued at the outset. One French government White Paper identified outdated registries as pervasive throughout the developing world:
[T]he laboriously created [registries] are often inoperable because they [are] rarely updated, either because there are insufficient resources to do so, or because this task is largely overlooked. A land registry is only relevant if changes (inheritance, sales, gifts, etc.) are recorded on a regular basis, otherwise it will only be a few years before the discrepancy between the actual and the documented situation becomes a new source of insecurity.Footnote 43
In Cambodia, the ensuing problem of inaccurate registries is exacerbated by bureaucratic practice and fraud because local land officials routinely rely on documentation rather than actual land use, which means that an individual with falsified documents can dispossess farmers who have peacefully possessed the land or have even registered it under now invalid systems.Footnote 44 A specific example of the problem created by this practice was documented in Special Rapporteur Subedi’s report on Botum Sakor National Park.Footnote 45 The land was granted as an ELC to the Union Development Group, causing the displacement of over 1,100 families, but the government only offered compensation to those who had land titles or whose names appeared in the family record book. The family record book system, however, often contains inaccurate information since families routinely fail to register new residents or subsequent transfers.Footnote 46 Greater community knowledge of the law and the need for registering transfers may help with some of these issues, but the experience of other poor countries and the failure of the Cambodian government to increase popular awareness of registration requirements do not provide much hope for long-term success.Footnote 47
Shortcomings notwithstanding, the registration process has had some notably positive effects. There is evidence that systematic titling has improved access to credit, as most banks will accept a registration certificate as collateral for a mortgage or other loan,Footnote 48 and possession or ownership certificates have proven useful in local disputes between community members of relatively equal status.Footnote 49 Another important consideration is cost: the cost per title of Cambodia’s systematic registration system is among the lowest in the world. Before one becomes too optimistic, however, the scale of the task should be kept in mind. The total number of land parcels in Cambodia is now estimated at upwards of ten million, so while 1.5 million systematically issued titles is impressive, to title the entire country will take another 45 years at the current rate.Footnote 50
3.2 Distribution
Since possession of land after 2001 is irrelevant for purposes of determining ownership, the only way to acquire new land, other than through registered purchase from a private party, is the distribution of state land through the SLC or ELC process.Footnote 51 Neither process is simple. Advocated at the time as ensuring legality and order, land must be formally registered as state land, confirmed as lacking a public purpose, and bureaucratically integrated into a specific concession before it can be allocated. Delays in the mapping and registration of state land, and a lack of bureaucratic transparency, have slowed and distorted the process and threaten to continue to do so.
3.2.1 Economic Land Concessions
An initial issue with ELCs is the opacity, at least to the public, of the process: “What has been agreed commercially in dozens of deals in every investment sector is regarded by Phnom Penh as confidential, despite the fact that the government is often selling or leasing public assets.”Footnote 52 On the other hand, the private sector complains that connected individuals move onto land they know is about to be granted in an economic land concession so they can gain relocation compensation. Another indicator of corruption, or at least strategic thinking, is that ELCs are apparently politically timed. Many concessions have been issued immediately following an election, but few before, implying that they are timed to avoid the negative press that frequently accompanies them. The degree of irregularity, however, is difficult to pin down. One human rights activist complained that the lack of adequate notice and an effective means of opposition meant that the government could be indiscriminate in granting land to private companies when the affected community was too small to be politically significant. A representative from the private sector, on the other hand, voiced concerns that haphazard media coverage “painted all companies in an area with the same brush” and deterred companies from investing in an area for fear of receiving unwarranted criticism based on the actions of its competitors. A lack of consistent bureaucratic enforcement and the risk of unjustified negative press can lead Western investors sensitive to issues of corporate social responsibility to avoid the Cambodian market entirely in favour of Vietnam or Thailand.
3.2.2 Social Land Concessions
Because the government did not promulgate the necessary regulations until 2008, there have been only a few years of experience with SLCs, characterized by one Land Ministry official as aimed at the “roots of homelessness.” In that time, however, progress has been made in establishing pilot projects as part of the Land Allocation for Social and Economic Development (LASED) programme. Over 1,500 households have received land parcels as of 2010 in Kratie and Kampong Cham province, and smaller projects have been undertaken in Kampong Chhnang and Kampong Speu provinces in co-operation with a Cambodian NGO. Local governments have also relocated 20,000 families from temporary settlements on state land in urban areas of Phnom Penh and near Preah Vihear temple.
The scope and manner of urban relocations, however, give reason for pause before declaring victory. Citing figures for the capital alone, the Centre on Housing Rights and Evictions (COHRE) estimated that an additional 70,000 people are threatened with forced eviction in Phnom Penh, and at least 150,000 people live in fear of eviction from their homes and land nationwide.Footnote 53 Since the Cambodian Constitution incorporates international law, the government has an obligation to provide adequate alternative housing and compensation for all those affected by forcible eviction, regardless of whether they rent, own, or simply occupy homes on the land at issue.Footnote 54 Eviction typically relocates residents of centrally located land in the capital to land on the outskirts, which often lacks basic infrastructure and services (e.g. running water and sanitation systems) and realistic access to education, health care, and employment.Footnote 55 The eviction process itself has at times become violent, for example in Sihanoukville when armed military and police forcibly evicted 105 families from Village 6 in Mittapheap District.Footnote 56 Therefore, while SLCs promise an orderly mechanism for respecting the housing, land, and property rights of poor Cambodians, in practice they can mean a significant reduction in quality of life. Indeed one NGO reported of SLCs that “not a single one has been completed in accordance with the relevant laws” and that they “have perversely been used to steal land from the poor rather than provide it to them.”Footnote 57
A recitation of the various deficiencies in implementation—families displaced from land they have cultivated for years, inadequate relocation facilities lacking basic services, valuable land given over to large companies in backroom deals—brings into question whether it really mattered whether Cambodia introduced the Torrens system or took a more incremental and locally sensitive path to land reform. If local governments choose to title only land where ownership is unquestioned, buyers and sellers even of titled land do not bother to register the transaction, the central government illegally favours multinational corporations and short-cuts the relocation process, and the courts are ineffective or unavailable to remedy legal wrongs, it is hard to imagine that the choice of one legal transplant over another would have made a difference. Yet it is possible that the increased national ownership of the processes of legal drafting and institutional implementation will eventually lead to greater accountability. As foreign influence recedes, one can hope if not necessarily expect, that the dynamics of land management will come to reflect the indigenous preferences of a country coming into its own after decades of trauma, but that process will at least initially depend less on the choice of institutional models than on the evolution of domestic politics. Given that governmental policy has been to start with the easy cases, with mixed results, optimism may be ill-advised, and the overall lesson may be that undertaking grand schemes without a clear understanding of the local context and consideration for entrenched interests of implementing actors is decidedly risky business.
3.3 Connecting Legal Reform to Policy Implementation
Humility should be the watchword, however, in making even a tentative evaluation of Cambodian land law reforms. Time has been short; reliable data are scarce; available reports often mirror the interests of their authors; and finding an appropriate metric may be impossible. Indeed the very goals of the land reform, integration into the international economy and tenure security for the Cambodian people, may point in different directions, and while judging domestic bureaucrats and foreign donors by the utopian rhetoric of proponents of land titling may be satisfying on one level, it suffers from the unacknowledged assumption of a contrafactual that might itself have had unanticipated negative consequences. The analysis that follows is offered with these limitations in mind.
The law and economics literature offers competing narratives regarding the shifting power dynamics between the individual and the state in the process of formalizing property rights. On the one hand, robust private property institutions theoretically increase the economic freedom and hence the political power of individuals, making them less vulnerable to the state.Footnote 58 On the other hand, the process of defining, surveying, registering, and titling formal property rights are invariably in the hands of state institutions, a situation that has led some observers to the conclusion that formalization gives the state “the opportunity to impose significant burdens on owners” while purporting to free the individual of governmental fetters.Footnote 59 The situation in Cambodia reflects this paradoxical world. The formal granting of ELCs to politically connected large industrial agricultural enterprises has undoubtedly increased their freedom from and power within and against the state, and it has done so with the rhetoric of transparent and secure property rights. Even if we postulate that the government’s reluctance to dispossess a powerful multinational is less related to law than to power and connections, the official celebration of the rule of law values gives a form of leverage to recipients, large and small, that an informal system would be hard pressed to equal. This power of legal rhetoric is particularly true in a country so closely watched by and dependent on the international community. Those outside privileged circles, however, face a very different experience. Not only may the relatively greater ability of powerful private entities to exploit legal formality further tilt the balance of power within the private sphere itself, but the state-driven titling process may also put them increasingly at the mercy of a state that may be incapable of vindicating their formal and informal interests, even when it attempts to do so.
An obvious issue is the need for informal payment at every step. While perhaps normatively less objectionable in Cambodian culture,Footnote 60 the result is increased cost and complexity, both of which work against effective administration and the enforcement of legal rights. Then there is politics. Local government leaders, even if honest, often acquiesce in large-scale land grabs because they fear state reprisals if they oppose politically connected outsiders.Footnote 61 Nor do judges or cadastral officials offer much hope. They are appointed along party lines, and political affiliation and prestige are often determinative of an issue. One critic quipped that “legal representation for the poor doesn’t really matter because it is all about whether you can pay the judge off. It might be useful to have a lawyer to make noise and publicize your case outside of the courtroom but it has little effect on the outcome.”Footnote 62
Of course “making noise” need not depend on a lawyer. Nor need the judicial result be the end of the story. Dispossessed Cambodians are willing to take matters into their own hands through protests to the ruling Cambodian People’s Party or symbolic marches to Hun Sen’s palace in Phnom Penh. “Noise” can work:
One community completely avoided the formal land resolution mechanisms and instead sent a complaint to the CPP party representative in their province asking him to resolve their land dispute. The Head of the CPP formed a team and met with the villagers. Following the discussion the community was verbally allocated 1200 ha (approximately the land area they had requested). While the verbal grant has no legal power it at least provides temporary security for the community and they saw results much faster than if they had gone through the Cadastral Commission.Footnote 63
And in Cambodia politics is not simply a domestic matter. In response to a 2006 corruption scandal over LMAP’s relocation of 4,250 families in the Boeung Kak Lake area of central Phnom Penh, the Centre for Housing Rights and Evictions filed a formal Request for Inspection with the World Bank. When the Inspection Panel discovered grounds for concern,Footnote 64 the political fall-out led to the World Bank’s withdrawal from LMAP in 2009.
It should be noted that the World Bank was not alone in devising the policy or guiding the implementation of LMAP, and activists criticized other donors including GiZ, Finnmap, ADB, and CIDA. The Bank was targeted because, unlike the other donors, it has transparent and user-friendly mechanisms by which outsiders can challenge project integrity. The success of the advocacy community is a credit to the World Bank’s responsiveness, and it may have consequences beyond the cessation of funding and the momentary political embarrassment of the Hun Sen regime. Although initially dismissive of the NGOs’ criticism and opposed to the Bank’s withdrawal, LMAP’s other patrons have become more conscious of social and political issues. GiZ’s parent organization, for example, has amended its terms of reference to include additional human rights protections.Footnote 65 There has also been discussion at the MLMUPC about instituting a “one window” programme wherein all the relevant authorities (e.g. the Ministry of Interior, the Ministry of Economy and Finance, and the MLMUPC) would be available in one place to simplify the process of obtaining an ownership certificate. Banking services would also be provided at this office to avoid the need to pay cash, and to increase the transparency of the fee payments.
More than most, the Cambodian government must respond to an external audience, and the court of international public opinion is likely more powerful and independent than any domestic one. High-profile evictions and protests over resettlement find their way into the media and reflect poorly on the government, which is often frustrated by its inability to match the activists’ adroit manipulation of the international media. Since the government is dependent on foreign aid for the continued functioning of many core public services, the NGO and donor communities are powerful players in domestic politics. “The Cambodian government should really employ a top notch public relations firm to manage its reputation on the international scene,” one private lawyer quipped. There are limitations, however, and they are not necessarily ones of diplomatic leverage. As the same lawyer pointed out: “The Cambodian government knows that the foreign donors won’t ever really pull out since so many of the foreign advisers have such a nice life here. The NGOs will only push so hard until they realize that their home country is actually considering withdrawing support.”
4. Conclusion
Cambodia may be unique in the degree of foreign influence and presence, but it is not exceptional in terms of land law reform in the developing world. Indeed, it is potentially an exemplar of the intersection of two global phenomena, both at least rhetorically intended to alleviate world poverty and hunger. The first is the global push to increase food production by facilitating foreign direct investment in developing countries’ agricultural sectors. The second is the ongoing effort by international agencies and donor countries to create “rule of law” systems in developing countries. These two have combined in Cambodia and elsewhere as foreign-inspired and financed land law reforms intended to increase the transparency of land ownership. Although the stated goals may include enhancing security of tenure for existing domestic users of land, as well as increasing transparency for outsiders, they are universally couched in the language of facilitating market exchange. In a process remarkably similar to the creation of colonial legal regimes,Footnote 66 complex localized patterns of usage are simplified into something akin to the concept of ownership imagined by Western legal systems since the English Enclosure Movement. Multiple users with fragmented rights are unified into a single entity with the power to exclude previous users and put the land into commerce; and the resulting interest is transformed into formal title and registered with central cadastral authorities so that the nature and value of the land and the identity of its owner can be ascertained cheaply, quickly, and conclusively. The immediate goal is to facilitate market transfers that will culminate in the land reaching its highest productive level, which will in turn achieve the ultimate goal of increasing economic growth and improving everyone’s wellbeing. Even the condition of the holders of customary interests unrecognized under the new system will be improved as the increased productivity translates into greater general wealth and higher social welfare.
Or so go the dreams of those suffering from “bright-line fever.” Ironically, however, the highest levels of agricultural productivity may be attainable by precisely the type of use threatened by the entry of land into international commerce. Agricultural economists have long argued that agriculture, unlike most forms of industry, is not scalable, and that the most productive form of farming is the family farm where the cost of both labour and its monitoring is lowest.Footnote 67 It is possible, therefore, that the conventional wisdom may fail even in theory, but the practical obstacles loom even larger. As long as the most common approach is universal centralized cadastral mapping, it is unclear that any reform could proceed without massive foreign aid and technical assistance. The use of satellite data to design cadastral maps may appear seamless when done by professional surveyors, but the software may not be so flawlessly responsive when the foreign expert returns to Australia or Denmark and is replaced by a rural Cambodian who has never owned an iPhone. Similarly a Torrens-style registration system may promise certainty in a developed society with an honest bureaucracy and a legally savvy population, but in a society where a bureaucrat can double his monthly salary by agreeing to misfile a document, or where a recipient of land from a relative or neighbour faces two days of travel and expenses equivalent to six months’ worth of crops to register the transaction, human factors may quickly erode that certainty. Even if we make all the necessary assumptions, waiting 45 years for registration to be completed while meanwhile refusing to recognize informal possessory and use rights seems a risky course of action that may further impoverish the rural population instead of enriching it.
Of course impoverishing a proportion of Cambodian farmers or urban residents does not mean impoverishing Cambodia. What may be expensive and confusing procedures to a poorly educated farmer may appear transparent and convenient to a foreign agri-business enterprise with professional legal assistance, and the misfiled or disallowed evidence of ownership that dispossesses an individual may be precisely the administrative move that secures the multinational corporation with central government connections exactly the tenure security that enables it to move the land into international commerce and thereby increase the national wealth of Cambodia, or at least the segment of society that benefits most directly from increased international contact and market fluidity and dynamism. In other words, the effectiveness of various approaches to legal reform may depend more on one’s metric—increased aggregate growth, egalitarian land allocation, security of tenure for its own sake, etc.—and perspective—that of an individual farm family versus a potential foreign investor or that of a World Bank economist versus an NGO rural activist—than it does on any agreed upon universal criterion.
As we stated at the outset, however, comparing the messy reality of a very poor country with a utopian dream of clear property rights readily exchangeable in perfect markets seems unfair even to those in thrall to the dream. It may only exacerbate that unfairness to replace the flawless top-down, technocratic utopia of formalization with a bottom-up romantic vision of rural communities with clearly articulable Geertzian “local knowledge” that will equally flawlessly lead to both maximum productivity and social justice. Neither utopia is likely to appear in Cambodia or anywhere else any time soon. It might be more useful to ask what the practical approaches might be.
While the 2011 legislation attempts to resolve contradictions between the Civil Code and the Land Law with some reference to the social context, the effect of the law remains to be seen. Furthermore, even if JICA scored a clear win over the World Bank team, many of the fundamental questions about top-down, foreign-led legal innovation will persist. What might be heuristically useful, therefore, is some speculation on what might have happened had Cambodia chosen to do nothing. Or, more precisely, if it had chosen to reject, politely, the wholesale introduction of foreign technical and legal expertise and attempted, undoubtedly with foreign money, to apply foreign experience selectively and with deference to whatever social and normative systems were (and probably still are) maintaining whatever degree of order and stability exists in Cambodian land practice.
An initial question, which we will only mention, is whether Cambodia would have been allowed to make this choice. The “rule of law” is a prerequisite to some of the privileges of developing countries in today’s world, such as favoured access to the US market, and there are legal criteria for entry into the WTO. While a panel of legal anthropologists might welcome a land law based on local practice, international organizations and the US Congress may be less sympathetic. And even if “utilizing its own resources,” as the former dean of Peking University’s law school proposes for China when he opposes mindless mimicking of Western legal models, is a possible choice,Footnote 68 it presumes that those resources exist and can be the foundation for law, and specifically for a law that will facilitate the market economy and foreign investment that Cambodia has decided it needs.
Perhaps it could. It is clear from the Chinese example that clear, legally formalized property rights and an independent and effective judiciary ready to enforce them against state intrusion are not necessary for economic growth or foreign direct investment. China has grown faster than any other country over the last 30 years, and is a leading destination for FDI, but that too may be a false comparison. What is working for China may not exist in Cambodia. The Cambodian People’s Party, for better or worse, is not the Chinese Communist Party, and the Chinese population at the beginning of reform, despite being abjectly poor in monetary terms, had many of the social prerequisites for growth. A post-conflict society, especially when that conflict involved Khmer Rouge rule, may not possess those prerequisites.
As the reader may have guessed, we are not going to attempt to answer the question of whether land law reform in developing countries is better facilitated by top-down social engineering by foreign technocrats or by allowing local residents to create indigenous systems of ownership and exchange incrementally. The question is unanswerable and the dichotomy is false, as Cambodia’s attempt to combine the technical precision of a Torrens system with the more flexible and open approach of the Japanese may eventually demonstrate. Nor does the fact that these systems are transplanted from other contexts constitute an indictment. Legal borrowing is ubiquitous in the contemporary world and may be the only way to build a legal system in Cambodia. It remains true, however, that any reference to foreign experience demands recognition that heuristic models usually not only presuppose levels of technical proficiency, social sophistication, and economic resources unlikely to be found in any developing country, but also that the models may reflect political preferences from a radically different time and place. Monetary compensation for a dispossessed victim of a registry mistake may have been an acceptable cost for the transparency and efficiency achieved by the Torrens system for twentieth-century Australians, but twenty-first-century Cambodians may have a different political and normative calculus. For the present, however, the most immediate fact is that Cambodia has made its choice, and it is too early to tell whether the choice will be a wise one or whether, as the observer quoted earlier said, it will not matter because the political and social realities of Cambodia will overwhelm any legal framework. Nonetheless, keeping these fundamental questions in mind may not be useless as we wait to see how the Cambodian land law story evolves.
Appendix A: table of abbreviations
ADB: Asian Development Bank
CIDA: Canadian International Development Agency
COHRE: Centre on Housing Rights and Evictions
CPP: Cambodia People’s Party
DANIDA: Ministry of Foreign Affairs of Denmark
ELC: Economic Land Concessions
FDI: Foreign Direct Investment
GiZ: Deutsche Gesellschaft für Internationale Zusammenarbeit GmbH
JICA: Japanese International Cooperation Agency
LASED: Land Allocation for Social and Economic Development
LICADHO: Cambodian League for the Promotion and Defence of Human Rights
LMADP: Land Management, Administration, and Distribution Program
LMAP: Land Management Administration Project
MLMUPC: Ministry of Land Management, Urban Planning, and Construction
MoJ: Ministry of Justice
NGO: Non-Governmental Organization
PRK: People’s Republic of Kampuchea
SLC: Social Land Concession
WB: World Bank
WTO: World Trade Organization
Appendix B: comparison of the 2001 Land Law And 2007 Civil Code
