Integrating land issues into the broader development agenda: Colombia^[44] - D.M. Grusczynski and C. Felipe Jaramillo

D.M. Grusczynski, Departamento Nacional de Planeación, Colombia
C. Felipe Jaramillo, World Bank

This paper summarizes the history of land issues and agricultural policy in Colombia, the resulting conflict over land use, and land reform programmes that title or redistribute lands designed to benefit campesinos (peasant farmers). The introduction of market-assisted land reform by Law 160 in 1994 is described in detail, as are two pilot programmes designed to incorporate market forces and local planning, and thereby improve beneficiary selection. The institutional issues that have led to past failures of land reform are discussed. The paper concludes by arguing for the need to implement the lessons from the pilot programmes and for a comprehensive rural land policy, which includes complementary measures that foster efficient land markets. Current armed conflict and fiscal restraints will make a large-scale land reform programme unworkable in the near future, but a peace package is likely to include a significant land access programme.

INTRODUCTION

This paper presents an assessment of rural land issues in Colombia with emphasis on continuing efforts to improve the distribution of property and provide land access for the rural poor.

Rural land policies and interventions have had a long history in Colombia. Policies and legislation have ebbed and flowed, according to the degree of interest in confronting tenure inequality or the problems of the rural poor. Land redistribution efforts were hotly debated in the 1960s when new legislation was approved, budgetary allocations were significant and social mobilization was intense. In the 1970s, redistribution efforts were replaced by a focus on integrated rural development. A new wave of debates on land reform efforts occurred between 1988 and 1994, when substantial changes were introduced to rural land laws. Most significantly, Law 160 of 1994 made possible a new market-assisted land distribution method; however, implementation and institutional problems have plagued its short history. Most recently, budgetary allocations for land redistribution have dwindled, despite widespread recognition that access to rural land and other interventions to help the rural poor are priorities on the development and peace agendas.

The changes in rural policy, and in particular land reform, have had little impact on the distribution of land or its use. Large tracts of arable land remain underexploited with subsequently low levels of rural employment generation, stagnation of rural poverty levels and environmental damage of environmentally fragile areas. Much of the policy debate has centred on subsidy levels and their costs rather than on the sustainability of productive projects. At the two extremes are (a) direct state intervention with high levels of subsidy for land and complementary activities, and (b) market-assisted land reform with lower subsidy levels in which the employment of credit to finance a portion of the land and complementary activities provides a market signal to the viability of the project. However, two pilot exercises indicate that beneficiary selection, planning at the local level and integration into the value chain are the key components of successful rural farm, and thus land reform, projects. Despite the successes of these pilot projects, institutional and policy failures have detracted from the results. Market failures remain and are reinforced by the continuance of, albeit reduced, subsidies and distortional agricultural lending policies. Equally critical is the failure of the land administration system to foster land markets, which together with armed conflict, have failed to guarantee tenure security. This paper takes the view that land policy should be addressed from a broad perspective that corrects the failures that impede effective land use, and that one component of this policy should include government support of access to land and production factors.

The paper is divided into five main sections. The first provides some background on rural land issues in Colombia, including tenure inequality, land underexploitation, the weakness of property rights and the increasing degradation of land and other natural resources. The second describes past policies and programmes that have affected rural land issues, with a brief overview of different generations of land reform. The third describes land reform policies in the 1990s, reviewing the content and implementation of Law 160 of 1994 and the pilot projects to develop the market-assisted land reform method. The fourth centres on the Productive Partnerships Support Project and other rural development policies of the 1998 - 2002 National Development Plan. The fifth section highlights key issues for the future including the rising prominence of land access, the debate on land distribution methods, the prevention of tenure concentration on the agricultural frontier, land taxes and the potential of land rental markets.

BACKGROUND ON RURAL LAND ISSUES IN COLOMBIA

This section provides some background on four issues related to rural land that are widely debated in Colombia and affect the poor. The twin issues of tenure inequality and land underexploitation are discussed together because these topics are often linked in Colombian discussions. The third issue is the weakness of the institutions that underpin property rights. Fourth is the increasing degradation of land and other natural resources. The section discusses each issue in turn, providing a brief summary of its relevance in the Colombian context, and an assessment of the importance of each.

Unequal land ownership and poor use of land vs its potential

Most of the attention on rural land issues in Colombia has been dedicated to the twin issues of high concentration of ownership and the underexploitation of agricultural lands. These issues have been traditionally linked because land that is kept idle or under low-intensity activities often forms part of large holdings that exacerbate the lopsided distribution of land. Furthermore, both issues have often been blamed as significant underlying causes of rural violence. Underexploitation, high concentration and the need to take action to generate jobs and improve the living standards of the rural poor have justified most public legislation on rural land issues and fuelled substantial government intervention since at least the 1930s.^[45]

FIGURE 1
Land cover and farm size

With a Gini coefficient of around 0.77 (Deininger in Castaño, 1999), Colombia's rural land ownership pattern is very unequal. Although this value is not unlike those for other Latin American countries, it is in sharp contrast to Asian economies, which have grown at rapid rates in the past decades, such as the Republic of Korea (0.35), Taiwan Province of China (0.45) and Thailand (0.45).

Although high inequality has often been used as sufficient ethical justification for redistribution, in the case of Colombia it is often linked to the underexploitation of a large proportion of rural land. A significant amount of good arable land in Colombia is held in large estates that are either left unused for long periods or that sustain low-intensity production activities such as extensive livestock grazing. Many studies have identified the inefficient pattern of use of land resources as a cause of the low rate of employment generation in agriculture, which marginalizes segments of the population and fosters migration into environmentally fragile areas such as steep hillsides and tropical rainforests (e.g. Heath and Binswanger, 1996).^[46]

Evidence of underexploitation is reported by the National Agricultural Survey (2001), which finds that plots of less than 20 ha that account for 14 percent of farmland^[47] comprise 40 percent of cultivated land. Parcels of between 500 and 1 200 ha comprise 20 percent of the agricultural area, but only 0.62 percent of cultivated land. Figure 1 shows that as the size of the plot increases, an increasing proportion of land is dedicated to livestock production.

Aside from underexploitation, surveys of land potential and use conducted by the Instituto Geográfico Agustin Codazzi (IGAC) suggest that a large share of rural land in Colombia is not dedicated to its optimal use. Their 2002 survey states that 62 percent of the land that has been affected by human activity is not used according to its capacity.^[48] Nearly 18 million hectares are underexploited (mainly used for extensive livestock grazing). The overutilization of nearly 20 million hectares includes land that would be best left to conservation or forestry, but is currently in use for crops or livestock grazing. The report goes on to state that nearly onequarter of grazing land is located on prime agricultural land, and over half of pasture land is located in areas for which conservation is recommended. Conversely, 71 percent of the land that could be dedicated to annual crops is currently dedicated to livestock grazing. Of the land dedicated to crops, only half fully exploits its potential.

The reasons for the concentration/ underexploitation conundrum are multiple. They include historical legacies, the many pecuniary and non-pecuniary benefits that larger farms enjoy, as well as the impact of agro-ecological diversity.

Historical legacy

A great deal of the inegalitarian pattern of land tenure in Colombia - as with other countries of Latin America - can be traced back to the dispossession of the native population by Spanish colonizers, the assignment of land in large estates to influential families and the postindependence pattern of repaying military leaders of civil wars with large land grants. In this way, much of the prime agricultural land was carved out very early on by colonial and governing elites. The pattern of frontier settlement of the late nineteenth and early twentieth centuries led by small groups of colonizers did little to change the pattern of land tenure as smallholders eventually gave up their lands to wealthier consolidators (Legrand, 1986). In the process, poorer families were displaced to poorer quality smallholdings, often on the hillsides of fertile valleys or further on the agricultural frontier.

Large haciendas were able to survive by tapping cheap labour in the form of tenants who were obligated to supply labour. As this mode of production broke down in the early part of the twentieth century, large estates should have disintegrated and smaller farms prevailed. Although the largest haciendas were in fact broken up, usually through inheritance or foreclosure, the resulting division has made little impact on the overall distribution of land because many new large estates have appeared, many in areas of relatively new settlement such as the Eastern Plains (FAO, 1994).

Economic distortions

In a country of rural labour abundance and low rural labour wages, smaller family farms should have had an obvious advantage over larger holdings that require employment of many workers at harvest time.^[49] However, many of the large holdings survived the disintegration of the hacienda system by transforming themselves to take advantage of a number of critical benefits unavailable to small farms that neutralized the diseconomies of the large estate.

Several studies have identified a number of key advantages that have benefited disproportionately owners of larger landholdings in Colombia. First, land has been a useful inflation hedge throughout much of the twentieth century, most significantly in inflationary periods when deposit rates in formal savings markets were capped (until the mid-1970s). Second, investments in large tracts of rural land have traditionally provided significant tax advantages. Appraised values have tended to be low, as have taxation rates. Third, rural land has been an excellent investment vehicle for investors who want to hide wealth, evade or avoid taxes and launder illegally obtained funds. Primitive recordkeeping systems in official registries do not allow the national tax agency to verify information on income-tax returns, making land an attractive asset to mask undeclared earnings. This is one of the reasons why there is a tendency for a substantial share of repatriated illegal drug profits to be invested in rural land. Hence, from the point of view of investors, large rural estates can yield significant benefits without engaging in agricultural production.

For those interested in farming, owning large estates has also yielded important advantages not available to smallholders. Better educated and organized farmers of large estates have always had an edge over their smaller competitors in obtaining access to formal credit sources.^[50] They also enjoyed privileged access to the government programmes established since the 1940s to spur the modernization of Colombian farming practices, such as subsidized loans, mechanization subsidies, marketing supports, and research and extension (García and Montes, 1988). Although many of the advantages related to government programmes and subsidies have been eliminated since 1990 with the implementation of liberalization reforms (Jaramillo, 1998), distortions remain or have resurfaced as incentives originally designed for smallholders were expanded to include large farms, or as "emergencies" have justified the creation of new subsidies.

A final category of benefits to owning land in large tracts is non-pecuniary privileges that are often mentioned in the Colombian debate. Difficult to quantify, these benefits refer to a degree of social prestige and status that is of great value to large landholders in Colombia.

Diversity of production

Another important source of heterogeneity in farm size in Colombia stems from the tremendous agro-ecological diversity of the country. As a result, flowers can be produced efficiently in "large"-scale operations on only a few hectares and yet require high levels of capital investment. By contrast, cattle farms in the Eastern Plains can run into thousands of hectares and be worth less in terms of agricultural returns than the cattle that graze the land. Cereal crops tend to be produced on medium to large holdings with low labour input and a high degree of mechanization. Natural advantages coupled with research and adaptation of technologies have facilitated large holdings under certain crops (banana, oil palm, sugar cane) and smallholdings in others (coffee, vegetables).

The coexistence of large farms alongside medium and small farms, as well as the preponderance of extensive cattle grazing activities have fuelled a long-standing debate among observers about the ideal farm size for Colombian conditions. Most empirical studies have shown that, generally, small farms tend to be more productive than larger ones, in the sense that they yield a higher level of total factor productivity (e.g. Berry, 1973; Forero et al., 2002). However, it has also been shown that large plots are efficient for a number of crops such as sugar cane, banana and oil palm, in which the timing between harvesting and processing are critical and important economies of scale exist for the processing facilities (Binswanger, Deininger and Feder, 1995). These crops, as well as those that utilize high levels of mechanization, were emphasized during the Pastrana administration, leading to the assumption by policy-makers that large farms are always more efficient, despite a lack of empirical analysis to indicate the general viability of larger farms without access to pecuniary and non-pecuniary benefits.

Land markets

There is a very active land market in rural Colombia, by some estimates more active than in the United States.^[51] However, the market is highly segmented. Large buyers purchase land from large sellers whereas small parcels are bought from small buyers. Large buyers do often purchase from smallholders on the agricultural frontier. Opposing transactions, involving the breakup of large estates into small ones through sales or rentals, do not usually occur, as would have been expected but for the privileges that owners of large holdings enjoy. The fact that large farms in Colombia are not sold in small parcels has been attributed to the value of the farmland being higher than the capitalized value of farm profits (World Bank, 1996).

The market for rentals is also highly segmented, with large commercial operators renting out from large landowners. The old hacienda pattern of small tenants sharecropping in large properties disappeared gradually after 1930 as tenants were expelled, as a result of legislation that gave rights to tenants and policies to promote farming in medium and large plots. In 1968, legislation was passed that made the existence of lease arrangements sufficient grounds for expropriation, accusing them of being inherently exploitative (World Bank, 1996). Sharecropping was banned with the Sharecropping Law of 1975. The result of these policies was an increase of 14 percent in the area worked by owners between 1960 and 1988 (from 77 percent to 91 percent). The amount of land cultivated through rental and sharecropping arrangements declined from 21 percent to 5 percent during this same period (Lastarria-Cornhiel, 1998). These provisions were reversed partially with Law 30 of 1988 and Law 160 of 1994, and rental and sharecropping agreements no longer accrue land reform benefits to renters or sharecroppers.

The changing rural setting

The evolution of land markets and use patterns in the past decade should have been affected by shifts in production brought about by falling commodity prices, efforts to increase the effective rate of property taxes and the intensification of the rural conflict.

Decentralization reforms encouraged many local governments to look on municipal taxes^[52] as a key source of income. At the same time, national efforts to improve valuation, mapping and registry systems have facilitated an increase in the effective rate of land taxation, despite the fact that there is still a large dispersion of land tax rates across municipalities. Although municipal tax reports do not differentiate between urban and rural taxes, the feeling is that collection rates for the latter remain low, thus failing to encourage the sale of large underutilized estates.

Additionally, expansion of the areas of conflict between guerrillas, paramilitaries and the Colombian army has introduced security risks, which have greatly affected farming in many areas. In some cases, holders of medium and large estates have opted to continue to run their operations from urban centres, leaving foremen in charge. In more extreme cases, farmers have sold their lands or abandoned them under threat; however, this has not necessarily translated into an increase in the supply of rural properties, as many landowners prefer to hold on to their properties rather than sell at a loss.

Since the early 1990s, a growing number of poor rural families facing threats from armed groups have been uprooted from their home areas, and today over 1 million attempt to eke out a living in urban centres in Colombia. Although it is unclear whether smallholders are displaced so that armed groups may gain tenancy to their land or simply as a strategy to gain territorial control, Kirchoff and Ibañez (2001) found that smallholders are more likely to receive threats than landless campesinos as are those who participate in community organizations. These threats, in turn, comprise one of the main causes of displacement, 58.2 percent of the displaced reporting to have received them. Displaced persons identified paramilitaries as being the main group that provoked displacement (58.7 percent) although guerrillas also provoked a large share (27.9 percent); however, more recent data indicate that guerrillas are increasingly displacing rural populations.^[53]

As a result of some of these trends, land prices in many areas of the country fell sharply from the high prices of the late 1980s and early 1990s as many owners looked to sell in order to repay financial obligations or leave the sector altogether. Notwithstanding this, in some areas of the country, forces have acted to prevent the breakup of large estates, in some cases further consolidating rural property. One of these driving forces has been the continuous need to launder drug money, which has sustained demand for large tracts of land. There is also anecdotal evidence of farm consolidation in some areas of the country where paramilitary groups have gained dominance since the early 1990s.^[54] Apparently, this is part of a political and military strategy to consolidate rural land and control territory in their areas of influence. Once paramilitary dominance is established, land prices tend to rise in these "secure" territories.

Weaknesses of the institutions that enforce property rights

The institutions that underpin property rights, facilitate transactions, make land markets transparent and foster land taxation suffer from many weaknesses that are common in other Latin American countries. Most importantly, large- and medium-sized holdings tend to enjoy proper titles, which are formally recorded in land registers, whereas a significant share of land held in smallholdings, particularly in remote areas, does not have formal title. Lastly, when titles do exist to these properties, they are often not registered.

To transfer land, the process requires that a notary authenticate the transfer of title, a step that is commonly undertaken. However, the transfers are not always registered, especially when involving small parcels of land in remote areas miles from a registrar's office. The failure to update the registry means that the cadastre is not updated either.

Despite these relative weaknesses, it is hard to generalize in a country as diverse as Colombia. In many areas, the majority of smallholders hold title, a result of massive titling programmes undertaken since the early 1960s. Since its creation in 1962, the National Land Reform Institute (Instituto Colombiano de Reforma Agraria, Incora) has specialized in rural titling programmes, dedicating a significant share of its resources to this activity. However, a weakness of past Incora efforts was the heavy emphasis on titling with little attention being paid to registration, a problem that has since been remedied. A more recent difficulty for registration is faced when the registrar's office is located in a municipality controlled by an armed group different from that at the site of the parcel, because travel between these areas can be difficult.

Problems associated with land tenure insecurity are well known (e.g. Feder and Feeney, 1991). Farmers without title or security tend to invest less in farm equipment and land improvements, use fewer inputs and obtain less output per unit of land than those with title. Furthermore, farmers with properly registered titles have greater access to formal sources of credit. Farmers without a strong formal stake to a specific piece of land are more prone to overexploiting it and degrading the natural resource base because of uncertainty of formal ownership. In rural Colombia, lack of title to property exacerbates the problems of displacement of the rural population, who have no way to reclaim land should they wish to return or sell what land they have cultivated. Although Law 387 of 1997 provides a system to freeze property transactions of displaced persons whether they hold formal title or not, the problem of enforcement remains. Of displaced landowners, 68 percent were owners, 12 percent resided on collectively owned land and only 11 percent claimed colono status with no formal title to their properties (Pastoral Social, 2002). A colono settles unclaimed lands and lives by exploiting natural resources usually through agriculture and thus incorporates these lands into the national economy. By definition these lands are usually on the agricultural frontier (Incora, 1973; IGAC, 1986 in Tesauro plurilingüe de tierras de la FAO, Rome 2003).

Despite these problems, as was explained above, land markets (for sales and rentals) in Colombia are quite active. Clearly, the absence of title and/or recording of transfers do not impede land sales or rentals (Suarez Gómez, 1992). However, informal markets are not transparent, and the resulting allocation of resources is likely to be suboptimal.

Degradation of land resources

Degradation of land and the associated natural resources has been accelerating in Colombia. In some areas, it is reflected in the overexploitation of hillsides, erosion of topsoil, destruction of valuable forest and water resources, and silting of streams and rivers. These problems are accentuated in the farms of traditional campesinos, who struggle to maintain yield levels despite a degrading resource base (Heath and Binswanger, 1996). In agricultural frontier areas, resource degradation is reflected in the felling of valuable, often primary, forests, and the exploitation of fragile lands that have little long-term agricultural potential. The latter phenomenon is often also associated with the planting of illicit crops.

The degradation of land resources in Colombia is a result of the increasing numbers of rural families attempting to survive in marginal conditions and the absence of employment opportunities in rural and urban areas. Despite declining population growth in rural areas over the past decades, the resource degradation process has been fuelled by high levels of poverty among rural dwellers and low employment growth, resulting from a decrease in the profitability of many commodities, and a continued policy focus on mechanization. The consequence is rural unemployment that has nearly tripled in the past decade to 11.5 percent in 2002. Urban centres have not been able to absorb gainfully the large migrations from the rural sector, and employment generation in agricultural activities has been low by international standards since the 1940s, as a result of growth centred on mechanized agriculture (World Bank, 1994).

Deforestation and overexploitation of fragile lands has also been aided by the inability of the public sector to enforce fully its property rights over national parks and other state-owned lands. Families who settle within the boundaries of national parks or other public lands have usually managed to obtain formal title, owing to the ineffective defensive actions by government agencies that eventually remove the restrictions of forestry reserves to title this land to colonos cultivating the land.

IMPACTS OF PAST POLICIES AND PROGRAMMES

Since the 1930s, Colombia has had a history of state interventions focused on achieving a more egalitarian land distribution, diminishing rural poverty and increasing agricultural productivity. Although numerous land parcels were either titled or redistributed to campesinos (at a total cost of US$4.5 billion or US$35 000 per family [FAO, 1994]), these interventions have had little overall effect on the degree of land concentration or rural poverty.^[55] Land reform efforts began as a judicial endeavour in 1936, which established a land statute with Land Judges who would grant title and redistribute land. The failure of this legalistic approach to land issues law gave way in 1961 to the government-sponsored promotion of land redistribution and titling, in which the state committed to providing land, technical assistance, credit, marketing and other support required to facilitate the success of beneficiaries.^[56] Over time, new institutions were created to implement each component of rural development, weakening land reform programmes by limiting them to the redistribution of land either through titling or purchase. In 1994, the state looked to the market to help resolve the distortions in land allocation and promote more efficient small farm enterprises; however, land reform remained a stand-alone programme out-of-sync with the demands of rural enterprise. The following section details the redistributive aspects of these different phases of land reform. During this time, a majority of Colombian agriculture underwent substantial modernization and achieved considerable increases in productivity often with assistance of an agricultural policy that became progressively detached from land reform.

The widespread use of exploitative tenancy arrangements, rural labour unrest and concern with the underexploitation of rural land prompted the passage of legislation of the first land reform law in 1936 (Law 200). Although the stated purpose of the law was to redistribute idle land and protect the rights of tenants,^[57] in practice redistribution of prime agricultural land was rarely accomplished. Landowners resorted to large-scale eviction of tenants and squatters to prevent loss of land. The law was also unsuccessful in facilitating significant expropriations,^[58] managing instead to induce landowners to modernize their farming practices and increase agricultural productivity. The displacement of tenants contributed to the social tensions that erupted in a long period of rural violence known as La Violencia (1948 - 1964).

A new wave of concern for rural land issues culminated with approval of Law 135 in 1961, which created the Incora, a land reform and titling agency, which was given authority to expropriate and redistribute unproductive estates and title state-owned lands. The legislation was motivated by the desire to create a family farm sector that would capture both the equity and the efficiency gains that redistribution promised. Redistribution activity through expropriation gained some momentum through the late 1960s, spurred by an increasingly politicized campesino movement.^[59] However, before significant transfers of land were accomplished, a growing number of invasions led to a landowners' revolt (Chicoral Pact), which effectively ended the redistributive phase in 1972. Since then, the rate of land redistribution has declined dramatically and has been mostly limited to voluntary offers by landowners. The decline in redistribution was exacerbated in 1982 when purchase prices were liberated from cadastre values, which dramatically increased the cost of buying private farms for redistribution (Incora, 2001).

In the face of the relative stagnation of redistributive efforts within the agricultural frontier, with the support of various multilateral aid organizations^[60] Incora undertook a series of directed colonization programmes distributing baldíos (unclaimed state lands usually considered beyond the agricultural frontier) in the yet untamed south-eastern plains during the late 1960s and early 1970s (Marsh, 1983). To facilitate economic development, Incora also coordinated complementary investments in health, education, transportation, irrigation, animal husbandry and crop production, as well as subsidized loans to land reform beneficiaries. The programme faced budget delays, unforeseen environmental difficulties and coordination problems with Incora's staff in the execution of the many projects it was charged with. In response to the failure of directed colonization and integrated rural development, Incora was relieved of most of its responsibilities, leaving it with the tasks of titling and redistribution.

Starting in 1973, attention to rural poverty issues concentrated on the new fad of integrated rural development, with support from the World Bank and USAID. The strategy was aimed at modernizing staple food production in the family farm sector and included subsidized loans, technical assistance, marketing support, provision of social services and infrastructure, as well as research and extension efforts in campesino crops. Farmers living in targeted areas displayed large gains in yields as a result of adoption of new seeds and modern inputs. However, owing to falling product prices, by the late 1980s campesino income had only managed to maintain their initial income levels. These integrated rural development efforts were not focused on land reform beneficiaries who were often left without access to complementary investments or technical assistance.

A new attempt to spur effective redistribution efforts arrived with the approval of Law 30 of 1988, which modified Law 135 of 1961. As it became evident that land redistribution was insufficient to improve the welfare of beneficiaries, new mechanisms to provide integrated rural development were considered. Rather than return to the earlier generations of Incoraled rural development encompassing land reform, investment in rural infrastructure and technical assistance, an institutional mechanism through the Consejo Nacional de Política Económica y Social (CONPES; National Economic and Social Policy Council), gave Incora wider powers to coordinate complementary investments in infrastructure and other services in those areas where land redistribution beneficiaries were to be settled. Every year, the complementary programmes were the first to suffer when it came to budget cutbacks, and Incora, with its own difficulties in executing its budget appropriations, lacked managerial expertise to coordinate with programmes whose directors had other priorities.

In 1994, Law 160 made a drastic break with past practices by opening the way for market-assisted land reform. The idea was to allow potential beneficiaries to negotiate freely with landowners and to select the land they wanted. Beneficiaries could apply for a direct subsidy from the government to cover as much as 70 percent of the value of the property. They were also expected to obtain long-term loans to cover the remaining 30 percent as well as the farm project. Besides assisting landless campesinos in the process of obtaining land through the market, Incora retained the traditional power to redistribute through direct purchases.

The idea of granting direct subsidies to beneficiaries met with strong resistance from entrenched political interests and Incora staff. The former perhaps feared more transparent procedures would affect their economic interests, whereas the latter seemed to be suspicious of the market in general, preferring that the state plan and execute land reform as well as other elements of rural policy. As a result, lengthy discussions on implementation procedures dragged on for several years and redistribution continued at a snail's pace, using only traditional direct purchase methods. A more detailed description of the problems with land reform practices in the late 1990s and the implementation of Law 160 appears in the next section.

The general sense among Colombian experts of land issues is that redistribution efforts since 1962 have not been successful. As much is recognized by Incora in its review of 40 years of Incora-led land reform published in Colombia: Tierra y Paz (Incora, 2001). One reason is that budgetary allocations for land redistribution were never large enough to make a dent on concentration indexes given Colombia's firm commitment to compensating owners at market value.

There is also a sense that a large share of beneficiaries of Incora's land distribution activities have failed in becoming successful farmers and that many of them were not able to lift their incomes beyond poverty levels. Productivity levels of land reform beneficiaries are much lower than renters and do not improve over time, suggesting that land reform failed to target more productive farmers (Gonzalez and Deininger, 2002, unpublished observations). Most beneficiaries planted low-value crops with traditional technologies and faced continued financial difficulties. Cases are often cited of families that have been forced to rent, sell or simply abandon their land (e.g. Suárez and Vinha, 2003; Zapata and Arismendy, 2003).

The failure of Incora's beneficiaries reflects many long-standing weaknesses of the land reform process. Families have often been given land without having access to critical complementary services such as technical assistance, lending and productive infrastructure, much less adequate health care and education. Poor agricultural performance has often been associated with Incora's poor technical assistance. Projects have been designed from urban centres, without taking into account market realities and the capabilities of beneficiary families. In some cases, the quality of land redistributed has not been ideal, in others access to roads and markets has been lacking. Often, beneficiary families do not possess the basic skills required to facilitate their success as farmers.

Critics of Incora have accused the agency of excessive paternalism, prevalence of political criteria in the selection of beneficiaries and land purchases as well as corruption. The beneficiary selection and farm project development methodologies applied officially until 1994 (and unofficially thereafter) assumed that beneficiaries were passive subjects who did not participate in selecting land, designing farm projects or planning their future.

Incora has also been singled out as a poorly performing agency in terms of its reputation for corruption, poor management and the inadequate technical skills of many of its staff. Incora's national and regional managers are political appointees as are other top officials, who, at times, exhibit an appalling lack of knowledge in issues relating to land access. The short stays of these officials are marred by shortsightedness or, worse, corruption. At the opposite extreme are the mostly unionized civil servants, of whom 73 percent lack a college education, and more than 60 percent of these staff are over forty, nearing the retirement age for Incora's special pension system (McKinsey, 1998). A lack of performance incentives and repeated restructurings have reduced staff from a peak of 3 653 in 1988 to fewer than 1 400 in 2001. Despite the contrasting political positions of the managerial staff, whose interests are tied to those of landowners and Incora's union members' left-wing views of land reform, the two seem to have reached an implicit agreement by which funding is provided to purchase and redistribute some land despite excessive costs, and maintain union benefits.

LAND REFORM POLICY IN THE 1990s^[61]

Law 160 of 1994

Law 160 of 1994 sought to give beneficiaries a larger role in the land reform process, limiting the role of Incora to that of a facilitator and relying for the first time on markets as a conduit for transactions. At the same time, it provided a new framework to integrate the efforts of a variety of rural and agricultural government institutions by creating the National System of Land Reform (SNRA) in recognition that land alone would not provide a solution to the difficulties faced by landless rural labourers and colonists on the agricultural frontier.

The law contemplated the following forms of acquisition and redistribution of land: (a) direct intervention by Incora, (b) market-assisted and (c) expropriation.^[62] With the direct intervention mechanism, Incora acquires a piece of land to distribute among beneficiary families subsidizing 70 percent of the value of 1 UAF^[63] per beneficiary family.^[64] This was the predominant method of operation before Law 160 and was expected to be employed less intensively with the introduction of the market-assisted mechanism. However, the law is vague regarding the limits to direct intervention, and a variety of programmes qualify for this subsidy: private lands within indigenous reservations, large holdings within Peasant Enterprise Zones,^[65] estate purchases when negotiated land reform fails, land for women who are the head of households, and land to complete 1 UAF benefiting minifundistas.^[66] The regulation of Law 160 stipulates that the remaining 30 percent of the cost of the land be financed by Incora, contravening prior legislation.

With the market-assisted process, land is selected by a group of campesinos who negotiate the price with the landowner. If the beneficiaries meet certain criteria, Incora subsidizes 70 percent of the cost of 1 UAF per family, with the remainder to be financed by the beneficiaries through savings or a bank loan.

The concept of the UAF as the size of plots to be distributed given in land reform dates to Law 135 of 1961 and has been justified by the desire to create self-sufficient, prosperous family farmers. This idea has introduced a degree of inflexibility to the market-based method. Some have argued that beneficiaries should have the freedom to determine if they desired smaller farm sizes, a strategy that would still allow a reasonable family income if farm revenues are complemented with those arising from other activities (i.e. services, non-farm employment or remittances from family members). This has limited the options available to some potential beneficiaries and may have saddled some with undesired levels of debt.

Regulations for the market-assisted mechanism outline the following procedures for its application:

• Campesinos inform the local Incora office of the estate they wish to purchase, describing its general characteristics and conditions of the negotiations. They may also ask for Incora's assistance in facilitating the negotiating process.

• Incora verifies that the campesinos qualify for land reform and are creditworthy, and notifies the landowner of the campesinos' interest in negotiating the estate. The landowner then informs the Incora whether they are interested in pursuing negotiations.

• Incora officials visit the estate to evaluate its agrological and environmental characteristics and calculate the size of the UAF. The landowner must produce title and other basic information about the property.

• Once all requirements have been met, and campesinos have reached an agreement with the owner, they sign the sales agreement, and Incora pays the subsidy directly to the landowner, with 50 percent of the value of the land paid in the form of low-yielding bonds and 20 percent paid in cash.

Incora can facilitate the process by distributing the list of properties for sale in its registries among campesinos, with information regarding the terms of the sale, and organizing meetings between landowners and campesinos. Should negotiations fail, Incora may proceed to acquire the property through direct intervention or even expropriation.

After about eight years of implementation, most market-assisted land reform has been Incora-led. Many potential beneficiaries have not been able to assimilate the cumbersome processes described in the regulations, leaving the decision-making and negotiating to Incora staff, many of whom have never fully embraced the market-assisted concept.

In spite of Law 160's efforts to emphasize the market-assisted method, budgetary allocations for this method were significant only between 1996 and 1998. Since then, resources allocated to land purchases have dwindled, along with interest in land reform. Since 2000, funds have been equally split between direct intervention and market-assisted land reform (Figure 2). The recent allocations for direct intervention reflect the need to provide land to displaced populations, who have also benefited from specific land policies.^[67]

The market-assisted mechanism was designed to rely on banks rather than government bureaucrats to review land quality and the feasibility of agricultural projects, as well as keeping land prices in check. Law 160 also required an additional price control, an absolute limit to the subsidy to be established by the Board of Directors of Incora. Neither of these mechanisms has worked in practice, and the price of the UAF has varied widely, ranging from a low of US$2 000 to US$22 000. This high, although within the limits established by the Board, required excessive levels of debt that no farm project could hope to repay.

FIGURE 2
Budget resources allocated to direct intervention vs marked-assisted land reform. Figures are in nominal terms

Since the market-assisted process has been in effect, the state-owned agricultural bank (Caja Agraria) has made nearly all the loans to land reform beneficiaries at the request of the Ministry of Agriculture and without regard to the cost of the land or feasibility of the project. However, with the closing of Caja Agraria and its replacement with Banco Agrario in 1999, this support has dried up. The new bank does not lend to land reform beneficiaries because it requires a 130 percent collateral guarantee, and does not accept land as collateral.^[68] Financing market-assisted land reform without credit has meant that either landowners finance 30 percent of the cost of the land, which they later write off, or sometimes they simply write off 30 percent of the cost immediately.

Another important section of Law 160 that has not worked in practice is the system for complementary investments. The SNRA was created to plan and coordinate policy and direct investment in land reform programmes. Its 27 members and six subsystems represent six ministries, the Department of National Planning, three of the government's agricultural institutes, presidential programmes, campesino organizations, indigenous organizations, agricultural suppliers as well as private agricultural associations. As might be expected, this cumbersome system never achieved integrated land reform. Even in the early years when some budget allocations were made by participating entities, these were the first to be eliminated by budget cuts (an annual event), because these entities and their directors often had commitments that differed from those of Incora. With no presidential mandate to support land reform, the system met only sporadically, and it no longer bothers to meet.

One problem that has not been overcome with Law 160 has been the continued focus of land reform efforts in areas where campesino enterprises have little hope of being successful. The law requires that priority areas for land reform be defined taking into account indices of unsatisfied basic needs, property concentration and rurality. The subsequent regulation has led to land reform efforts usually focused on municipalities with the lowest rankings without regard to land quality or distance to markets.

Market-assisted land reform and the pilot project

The Market-assisted Land Reform Pilot was developed in coordination with the World Bank in an attempt to demonstrate more effective methods to carry out the land reform process under the general guidelines of Law 160. In contrast with the excessive detail or omissions in the regulation of the legislation, the pilot project developed systematic procedures for market-assisted land reform. The most innovative procedures can be categorized in four areas: (a) selecting of beneficiaries capable of making a success of land reform, (b) planning land reform at the local level, (c) increasing the supply of land for sale and (d) developing the productive project in coordination with land reform beneficiaries based on the characteristics of the selected parcels.

The most significant problems detected in the implementation of the pilot were: (a) lack of financing for the 30 percent of the price of the land and the farm project, (b) the multiple rules to apply for government incentives executed by a variety of institutions and (c) guaranteeing the purchase of farm output.

The pilot project was implemented in five municipalities, which included an area of recent colonization, a mining town, a fishing village, an area dominated by oil palm plantations and a municipality on the peri-urban fringe of a provincial capital. Land concentration varies greatly among the different municipalities, as does water supply.

Law 160 requires that the selection of beneficiaries consider age, the number of children, connection to the region, agricultural experience and assets. With regard to assets, the point system favours those with the least capacity for implementing a successful farming enterprise by assigning more points to those with the smallest asset base. Although this achieves the redistributive effects of land reform, given the restrictions to financing complementary investments, a lack of assets may undermine the farm project.

In Incora-led market-assisted land reform the application review committee comprises Incora officials, representatives of local chapters of national campesino organizations and representatives of the candidates; however, they do little more than rubber stamp the list of beneficiaries. In the pilot project, the presidents of community action boards helped promote the project, while potential beneficiaries, Umata staff (Municipal Agricultural Technical Assistance Office) and municipal treasurers who administer the cadastre evaluated applicants. In all pilot projects, numerous applications were received. However, many of the forms contained false statements related to property ownership, residence and agricultural employment. A cursory review reduced the number by two-thirds. The pilot team proposed further selection criteria to narrow even further the list of beneficiaries, the most controversial of which were an upper age limit for land reform beneficiaries to rejuvenate the countryside and the requirement that at least one family member have a primary education, deemed necessary for business success. Although these criteria are less than egalitarian, the project team emphasized that the government must develop different poverty alleviation programmes for each population. As such, the pilot team, campesinos and their representatives, and Incora agreed to an age limit of 55 for the legal age of retirement for women, rather than 50 initially suggested by the pilot team. Enrolling applicants with insufficient education in adult education programmes helped mitigate controversy as well as pave the way for these beneficiaries to become future landowners.

After application reviews, the second step involved training and workshops for potential beneficiaries in which leaders and abilities were identified by the campesinos, and the responsibilities and the challenges were fully understood by each participant, fostering their commitment to the farm project or determining their withdrawal. Concurrent with training, local planning exercises were conducted, properties were evaluated and campesinos developed production projects. The training demands of the pilot project far exceeded those of Incora-led land reform and were considered by critics as too time-consuming. However, in a later survey of land reform beneficiaries, the majority responded that they would be willing to dedicate one day a week to training that taught them to manage their farm more effectively.

Given the rather limited local planning and the even more limited rural planning, the project team mapped out the municipality with the assistance of local officials and organizations.^[69] This identified areas of conservation, water resources, areas susceptible to natural disaster, the peri-urban fringe, infrastructure, indigenous communities and, finally, areas in which land reform projects could be implemented. The structure of landholdings and areas unincorporated into the cadastre were also reviewed.

Although the mapping out of a municipality may seem an obvious step in rural planning, such detail is not usually taken into consideration in land reform projects. Despite the intention of Law 160's framers to liberalize land markets, during the first two years of implementation of market-assisted land reform, only land registered with Incora was considered for purchase with little regard to its agroecological potential. The lack of planning in previous land reform programmes resulted in land being purchased for campesinos only to be later repurchased when indigenous reservations were established or enlarged. Similarly, urban expansion is one of the most common excuses used to sell land before the end of the mandatory 12-year holding period established by Law 160.

The regulation of Law 160 does consider agro-ecological conditions and specifically prohibits the purchase of land on steep hillsides. Although these rules may help avoid some poor choices, they still do not give the planners and beneficiaries the criteria for selecting plots that have the highest probability of agricultural and economic success. These strict rules simply cannot take the place of a participatory planning process, as demonstrated by the pilot projects.

In these projects, land reform candidates and local officials reviewed the maps assisted by a technical team. Large areas were determined unsuitable for agriculture, and the suitable lands were limited on average to about one-fifth of the municipality. This exercise eliminated from consideration most land in the registries of Incora, a fact that caused many problems for the pilot team, and eventually was a contributing factor in the rejection of the market-assisted land reform methodology.^[70]

Once the zoning was completed, the supply of land had to be identified. Because Incora's registries had proven inadequate, the project team helped the community identify underused estates whose owners might be interested in selling. Sources included lists of farms foreclosed upon by banks and the registry of deeds. Owners often lived in urban centres and were interested in selling their land although wary of becoming involved with Incora. The land was then subjected to soil and environmental analysis, and the farm plans were adjusted.

The farm project was developed concurrently with land and beneficiary selection rather than as an afterthought or a mere prerequisite to purchase land. One activity was to conduct an inventory of the skills of the land reform candidates to define possible productive projects. During visits to the potential parcels, farm projects were considered. The group of candidates then analysed the possible alternatives and determined which farm project or projects best suited each parcel of land. The analysis was based on business plans with financial statements developed by candidates. The productive project and the wishes of the beneficiaries were determinants in whether to keep the land in one piece or subdivide it.

The price of the land was negotiated based on its actual conditions, and not on estimates of future earnings of the farm project, which tend to inflate prices because they assume a variety of land improvements and market conditions that may not materialize.^[71] The initial selection of candidates as well as the self-selection that results from such a demanding process produced the final list of beneficiaries.

Comparison between the pilot project and Incora-led market-assisted land reform

Conceptually, Incora staff have traditionally seen land as the starting point of any land reform effort. As such, and often with the property to be purchased in mind, they identify beneficiaries, approving for practical effect the subsidy, and then design the farm project.

By contrast, the first step in the pilot project was local planning, followed by the selection and training of candidates, review of market conditions, development of the farm project concurrent with the selection of land and, finally, approving the subsidy that provides access to land and water. Both approaches to market-assisted land reform were charged with a variety of objectives, including: fostering rural business enterprises, using land according to its potential and permitting market forces to play a role in decision-making. To accomplish these objectives, Law 160 provided for the creation of land reform cooperatives, and detailed procedures for establishing the UAF and bank lending to finance the farm project.

The formation of cooperatives to manage the estate and farm project proved difficult to implement among many campesinos who had no prior relationship. Some cooperatives have been established to take advantage of the incentives offered by Law 160, but once the incentive is spent, these cooperatives have usually disintegrated. Another challenge for cooperatives is farm projects that depend on donated labour without clear sanctions for members who fail to give their share. These difficulties often lead to the formal or informal division of the estate into many individual parcels. The pilot team ran across similar problems, and the longevity of the cooperatives created in the pilot projects has depended on the relative success of the farm project. In a related experience, the flexible rules established during the pilot project to manage the community water well were a source of conflict when the campesino who owned the land where the well was drilled tried to charge fellow land reform beneficiaries for water consumption.

A small sample of both Incora-led and pilot projects in Rivera, Huila, indicate that both faced substantial obstacles in the implementation of farm projects. Land preparation requires working capital, and the gestation period requires that campesinos have a supplementary income source. In the pilot projects, funding was obtained for the preparation of land, the only income received by the campesinos during their first six months as landowners.

In both cases, the lack of financing for farm projects muted the success of the same. Notwithstanding, both marketassisted land reform approaches have resulted in significant increases in crop production and the reduction of cattle ranching. On one estate purchased in the pilot project, the area cultivated increased from zero to 47 percent. Less progress was made on property purchased by Incoraled land reform, but the increase from no crop production to 27 percent was still significant. In both cases, the cattle ranching that remained improved in quality.

Better farm projects reduced the cost and size of the UAF, which in turn resulted in lower debt levels. The cost of land as a share of total project costs ranged between 62 and 66 percent including land preparation in the pilot, whereas the cost of land in Incora-led farm projects reached 80 percent of total costs without land preparation. Similarly, the pilot project produced better results in terms of the cost of the UAF. In 1990, the UAF in Rivera cost $Col47.3 million per family, whereas in 1997 the pilot was able to purchase the UAF for prices ranging from $Col12 to 15 million. However, this was not only the result of better negotiation. The purchase of better land and the preparation of the same decreased the size of the UAF from 17 to 10 ha.

One of the major difficulties facing each land reform programme was the need for complementary investment. Given the failure of the SNRA to harness the multiple incentives offered by the national government, the project's field staff had to enlist the support of ministers and programme directors to provide complementary financing and bank loans. These efforts would have overwhelmed any group of beneficiaries. Incora staff also had difficulty finding incentives either through the SNRA or through informal channels, so it tended to focus on the incentive it controlled.

Although unable to address all the problems of land reform, the pilot project showed how planning significantly reduced costs of land transfer for the government while improving the results of land reform. The advances in market-assisted land reform were made possible by following the principle that rural development should be determined at the local level by citizens and local officials. Although the lack of municipal planning delayed the formulation and implementation of land reform projects, the reduction in costs is reason enough to justify the methodology developed by the market-assisted land reform pilot project.

The results of the pilot project were becoming known as a new administration was taking office. As the subsequent sections detail, this new administration was rather ambivalent to land reform, preferring to give rural policy a business focus. As far as small farmers were concerned, interest lay in making them competitive, modernizing their productive inputs and improving rural infrastructure. The committee meetings to supervise the pilot project were quite operational in nature, the staff who assisted on behalf of the ministry did not grasp the advances made, and with no bank loan in sight, Incora management lost interest in the methodology. Incora staff rejected the methodology, in part due to other studies conducted by the pilot project regarding the staffing of the institution, criticism of Incora practices, and attempts to gear land reform around productive projects and market issues rather than egalitarian principles. Although the methodology was presented to the Incora Board, it was never discussed or approved.

LINKS TO GOVERNMENT POLICIES, 1998 - 2002

With the failure of efforts to implement Law 160 successfully, support for land reform waned, as reflected in declining budgetary allocations. The Pastrana administration came in with little sympathy for land redistribution schemes and a desire to look for new ways to offer rural employment through the development of agricultural projects. As a result, the National Development Plan for the 1998 - 2002 period did not place land reform among its highest priorities, focusing instead on other policies to attract private capital into rural development through strategic alliances. The World Bank, eager to capitalize on the valuable lessons of the land reform pilot project, agreed to finance a new pilot that incorporated the new government's rural enterprise strategy into land reform projects. The government developed the Productive Partnership Support (PPS) Project, although it only reluctantly accepted that land redistribution issues be considered. The development of PPS yielded many valuable lessons about how to support development projects for the rural poor in the future. The lessons of these pilots also led to the creation in 2002 of the new Modular Incentive, which overcomes many of the limitations of previous government instruments to support the development of productive rural projects that may involve a land purchase or rental component.

National Development Plan 1998 - 2002

The three components of agrarian policy included in the National Development Plan 1998 - 2002 were productive value chains, rural development and institutional change. With regard to land reform, the plan promoted the market-based method as an efficient and transparent mechanism, but it emphasized the need for local planning of concrete productive projects. It also stressed the need to modernize Incora in its size, organizational structure and operations so as to guarantee efficient and flexible support of land reform. The plan also foresaw an important role for provinces and municipalities in targeting land reform efforts near rural infrastructure to achieve integrated development.

The plan reflected the government's preoccupation with the poor results witnessed over the years with successive land reform efforts. It suggested that peace and prosperity would not arrive to the countryside simply by distributing more land, emphasizing other methods to spur rural development, such as planning for rural development and discouraging the holding of unproductive land through taxation.

Where possible, the plan sought to involve private capital in rural development through strategic alliances between agribusiness and rural workers. The former would benefit by an increase in the supply of their primary inputs, and the latter would be integrated to the value chain, significantly reducing market risk. These initiatives were to be developed by the private sector in coordination with local communities. The government would act as a facilitator, financing activities with multiplier effects. The fostering of strategic alliances and productive nuclei was expected to assist in the reconstruction of social fabric, take advantage of successful productive experiences, create social protection for private investment and promote gradual access to land.

Productive Partnerships Support Project^[72]

The Pastrana administration did not pursue an expansion of the Market-assisted Land Reform Project because it did not consider land reform as the key driver of its rural development policy. Nevertheless, the lessons learnt in the pilot project were used to develop a new project, the PPS.^[73] The PPS Project fosters collaborative arrangements between organizations of small producers and agribusiness aimed at reducing technical, commercial, financial and/or social risks in pursuit of potential productivity and income gains in a particular value chain. Producers obtain access to markets for their products and critical inputs, while agribusiness firms expand food-processing activities by securing supplies from small producers that meet minimum quality standards. Although the government was not keen to foster land reform programmes, the pilot project did incorporate landless candidates into the pilots.

After much debate regarding where to start with the first pilot projects for the PPS, a project team^[74] in charge of its design decided to talk to business and identify potential firms that would be interested in supporting an alliance with a group of campesino producers. Although met with much scepticism by some industries, eventually seven agribusinesses expressed their interest in participating in the pilot programme (Table 1).

TABLE 1
The seven agribusiness participants in the PPS

*The partnership was fully developed, but not implemented.
**Security issues required that this partnership be suspended.

For each of the seven cases identified, the project team reviewed sector competitiveness agreements, company finances and expansion plans, and then visited the areas of the project and identified possible beneficiaries by inviting local campesinos and their organizations to discuss the partnership proposal.

Once local campesino organizations interested in the PPS pilot projects were identified, training focused on business and negotiating skills was offered. In subsequent workshops, campesinos and business representatives worked out the details of the partnerships. In six of the seven cases, local and regional universities and non-governmental organizations (NGOs) provided the training and facilitated workshops, significantly reducing costs while creating a regional knowledge base of the partnerships.

Based on a draft of each of seven project proposals, financial evaluations were conducted by external consultants in Bogotá who then made risk assessments. The results of the analysis enabled the project team to set strict limits to the total cost of land in the project. The consultants pointed out that the pilot projects included consistently lower costs for inputs than for other projects, probably as a result of local negotiations of costs rather than the adoption of standard farm projects.

Once projects were assessed and deemed viable, the next step was deciding what to do about land needs in each particular project. In the dairy, forestry and cocoa - tobacco partnerships, there was no need for land because campesinos were all previous land reform beneficiaries. However, many had unclear title or their land was at risk of foreclosure, issues that required correction before a partnership could be established. In the vegetable partnership located in Bogota's peri-urban fringe, members had inherited their land or rented parcels. By contrast, all three palm oil partnerships required the purchase of land.

Although the project team had resisted the initial inclination of the government to focus all projects in areas of armed conflict, rural violence led to difficulties in the formulation of the projects. Road travel for field staff resulted in delays as a result of blockades, and subsequent administrative problems regarding the strict rules for travel expenses that ignore the reality of the countryside. The imminent release of a paramilitary leader from prison led to massacres and massive displacement in the forestry project. The kidnapping of the agronomist from the organic production plantation and the subsequent emigration of all the directors of the family-owned business led to the collapse of this project. The assassination of a favourite mayoral candidate by paramilitaries frightened the workers of the Magdalena Medio palm-oil project, who decided they had little interest in owning land in the municipality in which they lived. Although violence is an issue that affects rural investment decision, it does not preclude all investment, and has only halted one pilot project. Agribusiness seems accustomed to certain levels of violence, and will take higher than expected levels of risk.

Land acquisition

The palm oil company from the Eastern Plains enquired into providing land for members of its labour cooperative through land reform, to which Incora replied that land could be supplied, but the company would have to work with a group of displaced persons with no previous experience in palm cultivation. Regardless, the municipalities prioritized for land reform by Incora regulation of Law 160 lay beyond the periphery of the potential nucleus for this palm oil-extraction plant.^[75] In spite of the restrictions placed by Incora, the partnership members began to define the areas in which to develop the project. Along with the criteria from the market-assisted land reform pilot, the distance from the plant and the political orientation of neighboring landowners, i.e. those with paramilitary leanings, were taken into consideration in the search for suitable land. Eventually a few parcels were identified, and the negotiation process began. Land was identified and the company financed this item.^[76]

The Magdalena Medio palm-oil project was located in an area dominated by cattle ranching and paramilitaries whose local administration was hostile to the plantation's expansion efforts. This hostility to the partnership proposal hampered the procedures outlined in the market-assisted land reform project manual with respect to detecting suitable land, and the project team was denied access to the registry office to learn the identity of owners of underutilized parcels.

At the suggestion of the project team, the palm-oil company accepted that former land reform beneficiaries dedicated to small-scale cattle ranching be invited to participate in the partnership. The company also contacted local landowners and proposed that the employees of the labour cooperative rent or lease the land. Unfortunately, no small- or large-scale landowner chose to participate in any partnership involving cooperatives, preferring to hold on to their rapidly appreciating land "protected" by paramilitaries with river and highway access to Caribbean ports.

However, the members of the cooperative were able to identify good land for sale within range of the palm-oil extraction plant, and they decided to request the market-assisted land reform subsidy offered by Incora. When the project team approached Incora regarding the acquisition of the land for the second palm-oil project, several problems arose. Because the parcel had not been cultivated in more than three years, Law 160 requires that it be expropriated. The project team was also aware that Incora staff disliked the partnership approach, which some believed to be a return to feudalism. The fact that the land was much less expensive and better than other land purchased by Incora also seemed to affect the decision not to purchase it. After several months of effort to persuade Incora to purchase the land, the team gave up and the pilot project eventually collapsed. To date, the land has neither been purchased for land reform nor have expropriation proceedings begun.

Modular Incentive^[77]

In 2001, as a result of the experiences with the PPS and market-assisted land reform pilot projects, a new instrument was designed to respond to the need for a flexible incentive that helps finance farm projects rather than only specific components. The PPS Project developed the Modular Incentive as a one-stop shop for small farmers participating in approved projects. Recognizing the holistic nature of establishing family production units in the context of productive partnerships, and that land is just one factor in agricultural production, the Modular Incentive may be applied to a range of production factors.

In contrast to land purchase subsidies, this incentive can cover the purchase of a range of inputs: (a) on-farm infrastructure, (b) durable goods, (c) vegetative materials, (d) operational inputs, (e) services, (f) labour - usually in the form of remuneration to the beneficiary or his/her family in the gestation period, and (g) land. The maximum value of the Modular Incentive is 40 percent of the required investment for a viable family production enterprise with a cap of approximately US$3 500 for beneficiaries who own land and US$8 500 for those who do not.^[78] No minimum was specified, thus allowing for smaller projects upon demand of beneficiaries and eliminating the potential rigidities of the UAF.

The procedures for the financial structuring of productive partnerships also take the support provided under the Modular Incentive as a means to close the financing gap of partnerships after other potential sources have been explored, rather than as a point of departure. In order to address concerns that incentives for rural projects oriented towards the poor might generate moral hazard in investment decisions, resources are required from the private sector and producers to complement financing. In order to emphasize this focus, partnerships are to be prioritized according to the amount of required support from the Modular Incentive per beneficiary.

The costs of the PPS

The precise costs per project are difficult to quantify because overhead, supervision, communications and office expenses were borne by government organizations; however, the project team estimates that each project cost between US$40 000 and US$50 000 to formulate. Direct expenses included field staff's salaries and travel expenses at wages double the average government salary. Training and the purchase of maps and diagnostic services (e.g. soil analysis and organic certification) accounted for a third of expenses; however, costs were kept relatively low by employing local organizations and professionals. Independent evaluations of the financial, environmental and social components that served to identify risks and improve project design comprised the remainder. As a percentage of total project costs, formulation accounted for less than 5 percent on average. Although centralstaffing costs should decline as the project progresses and local organizations are trained in the methodology, the pressure to reduce training and independent evaluation costs to derisory levels should be resisted. This initial investment in project formulation provides rapid returns through increased private funding (and decreased demand for government funding) of sustainable projects in a highly complex rural setting.^[79]

PPS risks

The success of the PPS Project in providing access to productive factors and converting smallholders into successful businessmen will depend on resolution of the armed conflict and the project management. National steering committees comprising government officials have been indifferent to innovative programmes or have quickly distorted them to suit their own needs. In the case of PPS, many of the procedures developed in the pilot phase have yet to be implemented. The key implementation problems revolve around project selection, and the lack of environmental, social, financial and economic evaluations prior to Modular Incentive approval. As such, the Modular Incentive does not appear to be closing a financing gap, but substituting other forms of credit. The high cost of land acquisition has led to a virtual veto of projects involving landless beneficiaries, and the Ministry of Agriculture has missed out on an opportunity to create competition for Incora.

The PPS Project will be executed through the Ministry of Agriculture, to create some competition to Incora-financed land reform based on the establishment of concrete farm projects integrated into the value chain. The project has been designed to avoid the centralized planning that has plagued Incora's projects, and local participants are meant to be the planners of partnerships. However, this decentralized design did not fully operate in the pilot phase, nor did the revolving credit account funded by incentive repayment, or the trust accounts designed to facilitate project management. These three components of the investment phase face risks that have hampered similar programmes: the failure to decentralize the programme and permit decision-making to take place in rural communities; nonpayment that has plagued agricultural lending; and the inability to maintain cooperative structures once government incentives are spent.

KEY ISSUES FOR THE FUTURE

The failure of past land reform efforts as a result of bureaucratic inefficiencies as well as corruption has dampened the interest of Colombians and their policy makers in traditional land reform efforts. The continued vulnerability of the fiscal situation will clearly limit the scope for land redistribution through the direct intervention land reform programmes of the past; furthermore, the high costs of marketassisted land reform and PPS may not provide a solution for a significant share of landless campesinos.

The rising prominence of land access issues

However, land access issues are likely to rise to prominence as Colombian policy makers attempt to cope with a stagnating economy, high urban and rural unemployment, continued rural poverty, and a protracted and complex rural-based conflict, which is generating a growing population of displaced rural dwellers. Furthermore, concern for water resources and the protection of Colombia's vast biodiversity will increase pressure from internal and international organizations to form a more coherent land policy. Any comprehensive strategy to solve the contemporary problems of Colombia is likely to include access to land among the most important issues.

Several factors suggest that greater access to land for a share of the rural poor and creative solutions to the rural unemployment problem will be required. One factor is that Colombia has been facing a slowdown in overall economic growth. Annual economic growth averaged 5 percent in the 1950 - 1980 period, but a structural break in the late 1970s has yielded average annual growth of only 3.0 percent between 1981 and 2000 (Cardenas, 2001). The growth slowdown has been accentuated since the late 1990s, with a sharp recession leading to a contraction in real gross domestic product (GDP) of 4.3 percent in 1999. The result has been an escalation of urban unemployment levels that hovered at around 10 percent in the early 1990s, peaked at 20.5 percent in 2000 and levelled off near 15 percent during 2001.

Another factor that is likely to increase attention to land issues is the stagnation of agriculture during the 1990s and declining rural job generation. Despite liberalization reforms in the 1990s, employment generation in the rural sector did not pick up and although some gains have been made since 1998 as the realexchange rate has depreciated, current prospects are not encouraging for the medium term. International commodity prices have continued their downward trend and most farmers in Colombia have had to look for ways to survive in an increasingly competitive environment by cutting costs.

In the meantime, the intensification of the internal conflict and the growing displaced population has brought rural issues to urban centres. The failed peace process with the Fuerzas Armadas Revolucionarias de Colombia (FARC; Revolutionary Armed Forces of Colombia) and the initiation of a peace process with the Ejército de Liberación Nacional (ELN; National Liberation Army) have brought land reform, albeit in the context of outdated leftist rhetoric, to the forefront of political debate. All these factors are likely to push Colombian decisionmakers to reconsider access to land as an important area of policy.

Some of the issues that are likely to rise to prominence during this debate will centre on the role of governmental command and control measures versus policies that rely more on market forces to increase access to land among the rural poor. The role of Incora in promoting land reform, together with its continued existence, will have to be reconsidered. Other issues for discussion include the debate on the role of land markets versus more traditional land reform efforts, the effectiveness of Peasant Enterprise Zones to avoid further land concentration, the use of taxes to break up large unproductive holdings and encourage intensive land use, and the potential of rentals as an alternative route to access land.

The debate on land transfer methods

Land markets are likely to continue to be at the centre of the debate to make land reform efforts more effective and less costly. Whereas supporters of Law 160 see the market as the most efficient vehicle to transfer land, others propose a return to a larger role for Incora in the purchase and redistribution of land. Critics of the market point to the ease with which land prices can be manipulated by large sellers and the disadvantaged negotiating position of small buyers. Payment of high prices to land sellers, coupled with high interest rates on complementary loans, has led to high indebtedness among beneficiaries, and the inability to inject fresh capital in farm projects has soured the marketassisted land reform experience. There is widespread perception that the experience with the "market" has been negative, just as campesino organizations and those of the left suspected. Academics also criticize the market approach as naive given the segmentation of land markets and the violence that dominates the countryside (Suarez, 1992). Scepticism of marketassisted land reform has fuelled many recent proposals to strengthen Incora's role as the main arbiter in the land reform process, to boost its budget and to allow it to pay values recorded in the cadastre rather than market values for land. However, others see the failure of four decades of land reform as justification for the closure of Incora.

Incora, an institution in crisis

Incora has been repeatedly restructured, a fact that has led to a staff reduction of more than half from its peak in the late 1980s. However, more recent restructuring efforts have been limited to the imposition of a hiring freeze on Incora and the elimination of vacant positions. Incora now is suffering from a shortage of technical staff and at times is without sufficient personnel in its regional offices to carry out basic tasks such as titling.

Total spending on land reform has declined in both nominal and real terms from a peak of $Col130 billion in 1996 to $Col68 billion in 2002. With land reform budgets in decline, Incora's operational, debt servicing and pension expenses now account for 73 percent of all costs. This decline compares with budget allocations in 2001 and 2002 of nearly $Col250 billion to the Investment Fund for Peace rural development projects.^[80] Although it did provide funding to title land to campesinos,^[81] the FIP considered that government incentives were best invested in making land productive rather than redistributing more land.

Declining budget allocations to Incora has raised the question of the Institution's continuance, but this is not simply an Incora issue. The numerous programmes and agencies that execute agricultural policy create confusion and may work to the detriment of agricultural policy. To add to the problem, the national government executes rural projects through agencies that directly report to the Administrative Department of the Presidency. Academics, DNP and the Ministry of Agriculture have all suggested that the agricultural policy can be more effectively implemented by reducing the number of agencies.

Most agree that, at the very least, Incora, the Fondo de Desarrollo Rural Integrado (DRI; Fund for Integrated Rural Development) and the Instituto Nacional de Adecuación de Terras (INAT; National Irrigation Institute) should be merged. However, this restructuring became entangled in legal debates on the scope of the President's special powers given by congress to restructure government, civil service rules, and which agency should accommodate the other two. Although some proposals were developed, these were never made official, and only minimal staff reductions were incorporated into restructuring plans.

Rural land policy in Colombia is executed by a myriad of agencies, many of which do not report to the Ministry of Agriculture. This is compounded by a political system that builds consensus around fleeting political appointments. The result is a collection of ministers and agency directors with conflicting and rapidly changing objectives who are unaccountable to the public. The failure of the political system to develop a land policy framework is best exemplified by Incora's poor management. Leaving aside issues of corruption and incompetence, decades of poor management has left Incora an institution in crisis. Its overstaffed Bogotá office reflects the centralized nature of its decision making and operations, and its lack of professionals render it unable to provide support in the development of sustainable farm projects. The elimination of vacant posts means there is little hope of rejuvenating the land reform programme. The organizational structure of Incora is at odds with the demands of an efficient land and rural development policy, but restructuring or closing the institution will not resolve management's failure to plan, budget, or monitor and evaluate performance.

The progressive reduction of land policy to a handful of land reform instruments since the late 1960s reflects the lack of initiative on the part of the Ministry of Agriculture, who have failed to consider complementary measures related to land taxation, agricultural incentives and institutions, and conservation policies that should be reviewed and articulated into a coherent land policy, thereby increasing the economic benefits for society of any land policy.

Controls on concentration through ZRCs

Peasant Enterprise Zones (Zonas de Reserva Campesina or ZRCs) were created by Law 160 in order to limit the size of holdings in areas of recent colonization. The application of ZRCs effectively limits land market operations in these areas. Their aim is to avoid repetition of the slash and burn cycle and subsequent expulsion for livestock grazing by creating buffer zones that protect natural resources and foster economic development through local planning. To avoid the concentration of land, landholding limits are to be established by the board of directors of the Incora. If holdings exceed the permitted size, Incora is authorized to acquire the properties through direct intervention.

Overall, the development of ZRCs has met with limited success. Of five ZRCs approved, three were included in the World Bank's ZRC Pilot financed with a learning and innovation loan. These three pilot projects have made some advances in the construction of communities through "learning-by-doing" projects financed by loan proceeds; however, activities have been skewed towards infrastructure and productive projects rather than local planning required to "consolidate the agricultural frontier". Nevertheless, these three pilot projects located in territories dominated by guerrilla groups have fared much better than the two Incora-led projects that are located in areas disputed by paramilitaries and guerrillas. The latter have received little financing for economic and social development, and it is not yet clear whether they will be able to prevent the displacement of the smallholders whom ZRCs seek to protect.

Land taxes

Land taxes have been the subject of much controversy in Colombia. Proposals have been periodically aired in national debates about the need to impose high taxes on land to force owners either to cultivate or to sell. With the interest in promoting market-based land reform after 1994, some attempts were made to boost taxation levels in order to foster a larger supply of farms to the market.

The collection of land taxes has not been a priority in Colombia until recently. Land tax rates have historically been quite low, cadastre values have tended to be far below market values and enforcement costs have been too high for the overwhelmed municipal governments of rural areas. Rapid decentralization efforts since the mid- 1980s have not created strong incentives for local governments to pursue land taxes as an important source of income. Transfers from the central government have been increased sharply and rewards for those who have made efforts to levy local taxes have only been put in place since 1998, as a result of the financial crisis of many municipal governments that became deep in debt. However, in 2000, transfers from the central government in its various forms provided 65 percent of municipal income, followed by a tax on industry and commerce of 11 percent, and property taxes of 10 percent (author's own calculations based on municipal tax collection reports to the DNP, 2000). Municipalities that collect property taxes charge on average three basis points (collecting even less), far below the 16 points the law permits.

Despite past discouragements, it is likely that land taxes will continue to be part of the future policy agenda, both as a potential source of revenue for rural municipalities and as a way to promote land sales and rentals. New investments in modern cadastre updating methodologies as well as lower enforcement costs are likely to facilitate collection efforts gradually over the next decade. Difficulties will remain in dealing with entrenched landowning elites who have been hard hit by falling profits and violence and have developed a strong defence mechanism by eliciting sympathy for their plight among urban constituencies.

Improving market access through land rentals

Another issue that will continue to be hotly debated is the role of rentals and sharecropping arrangements to provide expanded land access to lowincome campesinos. Law 160 repealed the remaining vestiges of previous legislation that outlawed sharecropping and discouraged the expropriation of land by granting landowners more rights.^[82] Although studies of sharecropping and rental arrangements recognize that these practices are slowly disappearing in the Colombian countryside, the fact that they exist at all after the many legal restrictions imposed indicates their necessity. However, warning must be given that these arrangements rarely occur between members of different social classes. As such, commercial producers rent from large landholders, whereas campesinos rent or sharecrop land of other smallholders as a means to complement the factors of production they lack. Although these arrangements between smallholders permit land that might otherwise become idle being put into production, they do not address the problems of large landholdings being underutilized.

Leaving land idle that could be rented out might appear irrational. However, the history of land expropriation by Incora, who later redistributed to it renters or (as was more often the case) squatters, has made many landowners reticent to become involved in any way with landless campesinos. The possibility that rented land may be occupied, the lack of cooperation by the authorities in ousting squatters and possible retribution by squatters have deterred landowners from renting out their land. Campesinos may also be reluctant to rent land from large landowners given the history of unfair contracts, failure of wealthy landowners to honour these contracts and the history of violence between the two groups.

Lastly, some large landowners simply have no interest in renting out land because their purchase was motivated by money laundering or other non-agricultural concerns. These properties, once consolidated, can be considered withdrawn from the market until long after they are expropriated (for acquisition by illicit means or funds), because campesinos are unwilling to settle for some time the land of violent criminals.

CONCLUSIONS AND RECOMMENDATIONS

Unequal land ownership and underutilization of productive lands are chronic problems in rural Colombia. They have been linked to an insufficient rate of employment generation in the countryside and continuing high levels of rural poverty. As documented here, the resolution of these problems has been at the heart of the bulk of rural land policy since the 1930s. However, the evidence reviewed suggests that the impact of government efforts in this area has failed to resolve the fundamental problems of inequality and underutilization. Despite several major legislative efforts, methodological changes and fiscal resources, both issues remain as intractable in the early 2000s as they were in the 1930s and the 1960s.

Despite the failures of the past, we believe that rural land access issues will continue to be a relevant policy issue for years to come. Although recent administrations have not put rural land issues at the top of their agenda, we are convinced that eventually they will need to be addressed. On the one hand, it will be difficult to develop an inclusive and pro-poor rural development strategy without facilitating land access to part of the landless population. On the other, it is unlikely that the rural conflict can be resolved without including a substantial land access programme.

It is difficult to propose decisive policy action for the short term while there is an ongoing complex war in which two guerilla movements, paramilitaries and the military fight each other on multiple fronts in the Colombian countryside. It is unfair to expect potential beneficiaries of land reform or productive projects to thrive while their basic security is under threat. The poor investment atmosphere in large areas of the country that has been generated by the conflicts suggests that government efforts in fostering new farm units should be selective and demand-driven, with priority given to relatively stable areas with productive infrastructure and market access.

Whereas some would argue that the government should launch an ambitious programme of land reform in order to diminish the causes of conflict, we believe that the current conflict involves much more than a struggle for land and that a costly reform process can only be justified if it is part of an agreement that yields true peace. While the conditions are ripe for such a possibility, we believe that government policy should focus on encouraging land taxes among rural municipal governments and undertake small selective projects following the PPS scheme that include small farmers and landless campesinos. The PPS pilot project incorporated many breakthrough elements into farm projects - such as business negotiation that provided access to factor markets, risk sharing between campesinos and agribusiness, and independent financial, social and environmental evaluations to improve the formulation of farm projects. The Modular Incentive could prove a very useful tool for such projects in the short term and we highly recommend that it be tested further. By contrast, pursuing broader land reform efforts, be it through the market or through a state agency, is not a realistic option in the current context, in part because of the high fiscal costs that it would entail, at a time when the government has embarked on tough retrenchment efforts.^[83]

Other actions that should be adopted in the short term include harnessing the many scattered rural development programmes under a few strategic objectives that fulfil the broad policy mandates of land reform and rural development. Such a merger of rural agencies would lead to fiscal savings and facilitate the coordination of policy objectives and implementation strategies. Given that institutional design does not guarantee the success of rural policy, the merger must go one step further and manage rural development based on results that permit the monitoring and evaluation of management and policy instruments to make adjustments as needed.

Additionally, clarification of property rights through titling and the modernization of the cadastre and registry are areas in which the involvement of government institutions outside the rural sector is required. The land rights of displaced persons must be protected through capable land administration systems, and coordination with organizations in the field that can carry out the required tasks.

The government should also be encouraged to advance towards the formulation of a coherent land policy, which reflects the articulation of complementary land tax, agricultural incentives and conservation policies. Such a policy should provide the framework for land and factor markets to operate efficiently, significantly reducing the cost per family of future integrated land reform efforts.

For the medium and long term, a good part of the solution to access and underutilization problems will need to rely on stronger systems of local land taxation and on the elimination of the privileges that have prevented many large farms from being broken up. Policies that artificially prop up this sector, such as mechanization incentives, soft loans, credit guarantees and subsidies for production and irrigation, should be eliminated.

BIBLIOGRAPHY

Anrup, R. & Renzo Ramírez, B. 1994. Aparcería y disposición en una hacienda cafetera del Tolima: La Aurora 1948 - 1993. Cuadernos de Desarrollo Rural No. 33. Pontifica Universidad Javeriana, pp. 9 - 42.

Berry, R.A. 1973. Land distribution, income distribution and the productive efficiency of Colombian agriculture. Food Res. Inst. Studies, 12(3): 199 - 232.

Binswanger, H., Deininger, K. & Feder, G. 1995. Power, distortions, revolt and reform in agricultural land relations. In J. Behrman and T.N. Srinivasan, eds. Handbook of development economics, III-B. Amsterdam, Elsevier.

Cardenas, M. 2001. Economic growth in Colombia: a reversal of fortune? CID Working Paper No. 83. Cambridge, MA, USA, Harvard University.

Castaño, L. 1999. La distribución de la tierra rural en Colombia y su relación con el crecimiento y la violencia: 1985 y 1996. Universidad de los Andes, Bogotá. (Masters thesis)

CRECE (Centro de Estudios Regionales, Cafeteros y Empresariales). 2001. Cuantificación de los impactos micro y macroeconómicos de la crisis cafetera. Manizales.

DANE. 2002. Encuesta nacional agropecuaria: resultados 2001. Bogotá.

DNP, SISD 31. 2001. Impacto Social de la Crisis: Diferenciales Urbano-Rural.

FAO and Centro de Estudios Ganaderos y Agrícolas. 1994. El mercado de tierras y la formación de propietarios en Colombia: Estudios de casos. Rome.

Feder, G. & Feeny, D. 1991. Land tenure and property rights: theory and implications for development policy. World Bank Econ. Rev. 5(1): 135 - 153.

Forero, J., Galarza, J.A., Torres, L.E. & Forero, J.L. 2002. La Economía campesina colombiana, 1990 - 2001. Cuadernos Tierra y Justicia Volumen 2. Bogotá, Instituto Latinoamericano de Servicios Legales Alternativos, ILSA.

García, J. & Montes, G. 1988. Coffee boom, government expenditure and agricultural prices: the Colombian experience. Research Report 68. Washington, DC, International Food Policy Research Institute.

Heath, J. & Binswanger, H. 1996. Natural resource degradation effects of poverty and population growth are largely policy-induced: the case of Colombia. Environ. Devel. Econ., 1: 65 - 83.

IGAC (Instituto Geográfico Agustín Codazzi). 1986. Situación y análisis del proceso colonizador en Colombia. Análisis Geográficos No. 10. Bogotá.

Instituto Colombiano Agustín Codazzi and Corporación Colombiana de Investigación Agropecuara. 1994. Zonificación de los conflictos de uso de las tierras colombianas. Bogotá.

Incora (Instituto Colombiano de Reforma Agraria). 1973. La colonización en Colombia, una evaluación del progreso. Bogotá, Incora.

Incora. 2001. Colombia: tierra y paz: experiencias y caminos para la reforma agraria, alternativas para el siglo XXI, 1961 - 2001. Bogotá.

Jaramillo, C.F. 1998. Liberalization, crisis and change in Colombian agriculture. Boulder, CO, USA, Westview Press.

Kirchoff, S. & Ibañez, A.M. 2001. Displacement due to violence in Colombia: determinants and consequences at the household level. Discussion Papers on Development Policy. Bonn, Germany, ZEF (Center for Development Research), University of Bonn.

Lastarría-Cornhiel, S. 1998. El arrendamiento de tierras en Colombia: prácticas y marco teórico - histórico. Colección Documentos de la Misión Rural: Documento 14. Bogotá.

Legrand, C. 1986. Frontier expansion and peasant protest in Colombia: 1850 - 1936. Albuquerque, NM, USA, University of New Mexico Press.

Llano R., Montoya, M.M. & Mora, L.S. 2001. Contraloría General de la República. ¿Incentivo a la Capitalización Rural? Políticas Públicas. http://www.contraloriagen.gov.co/publicaciones.

Machado, A. 1998. La cuestión agraria en Colombia a fines del milenio. Santa Fe de Bogotá, CEGA.

Machado, A. & Suárez, R. eds. 1999. El mercado de tierras en Colombia: ¿Una alternativa viable? Bogotá, Tercer Mundo Editores.

Marsh, R. 1983. Development strategies in rural Colombia: the case of Caqueta. Los Angeles, CA, USA, UCLA Latin American Center, University of California.

McKinsey. 1998. Apoyando en la Preparación y Ejecución de un Plan de Modernización: Diagnóstico y recomendaciones. Bogotá.

Ossa, E.C., et al. 1998. Análisis sobre la distribución de la propiedad rural de la tierra en Colombia 1985 - 1996 y los determinantes de los cambios. Working document.

Pastoral Social. 2002. RUT Informa sobre desplazamiento forzado en Colombia. Boletin Trimestral Nos. 14 and 15, April and September. Conferencia Episcopal de Colombia. Bogotá, Editorial Kimpres Ltd.

Pérez, E., Francisco González L. de G., Ricardo Dávila, R., Forero Alvarez, J. & los promotores de la Fundación San Isidro. 1986. Capítulo IV: La tenencia de la tierra en la actualidad. Cuadernos de Agroindustria y Economía Rural, No. 17. Pontifica Universidad Javeriana, pp. 65 - 104.

Raymond, P. 1986. La aparecería en las haciendas Charaleñas. Cuadernos de Agroindustria y Economía Rural, No. 17. Pontifica Universidad Javeriana, pp. 105 - 156.

Rojas, M. 1999. Una mirada institucional de la negociacion voluntaria de tierras rurales como estrategia de redistribucion y equidad. In A. Machado & R. Suárez, eds. El mercado de tierras en Colombia: ¿una alternativa viable?, pp. 67 - 104. Bogotá, Colombia, Tercer Mundo Editores.

Rojas, M.E. 2001. Evaluación de las Leyes 30/88 y 160/94 y diseño de indicadores para medición de la reforma agraria en Colombia. Bogotá.

Suárez Gómez, R. 1992. El mercado de tierras en Colombia: lo que no puede. Coyuntura Agropecuaria, 9(3). Centro de Estudios Ganaderos Agrícolas. Bogotá, Litobanganadero.

Suárez, R. & Vinha, K. 2003. Elementos para una reforma agraria efectiva. Base study for land policy: understanding the links between land distribution, markets and violence and the resulting land use in Colombia. Bogotá.

Urbina, H. 1999. Hacia la construccion de un mercado de tierras. In A. Machado & R. Suárez, eds. El mercado de tierras en Colombia: ¿una alternativa viable?, pp. 105 - 136. Bogotá, Colombia, Tercer Mundo Editores.

World Bank. 1994. Colombia, poverty assessment report. Washington, DC.

World Bank. 1996. Review of Colombia's agricultural and rural development strategy. Washington, DC.

World Bank. 2001. Colombia: Productive Partnership Support Project: project appraisal document. Washington, DC.

Zapata, J.G. & Arismendy, C. 2003. Ordenamiento Territorial y Reforma Agraria en Colombia: un Nuevo Enfoque para un Viejo Problema. Base study for land policy: understanding the links between land distribution, markets and violence and the resulting land use in Colombia. Bogotá.