February 1997




Fourteenth Session

Rome, 7-11 April 1997, Red Room



Item 5 of the Provisional Agenda





- 'Third Generation' Land Reforms 7-8


- Household activity choices under incomplete markets 12-14


- What kind of participation and access do land markets offer the rural poor? 18-20


- Effects on Productivity in Agriculture 24-26




1.New trends and models of agriculture are emerging. Increasing globalization, declining public sector budgets and the emergence of new markets has led to profound changes in the world's agricultural sectors, both at national and international levels. Many of these trends are reflected in the commitments and objectives of the 1996 Rome Declaration and World Food Summit (WFS) Plan of Action, and can be characterized as follows:

The Defining Trends of the New Agricultural Model

*    An extended agriculture, that transcends simple primary production linked to other economic agents in different forms of horizontal and vertical integration. This agriculture is highly dependent on effective and efficient services (Commitment 3 of the WFS).

*    A contractual agriculture that promotes partnerships and alliances among different productive agents, encourages private contracts (such as leasing of land and share-cropping) establishes links with the business community and with farmers' associations (Commitment 3).

*    A flexible agriculture. A focus on linked markets (land-credit, land-labour, labour-credit) is indispensable for adequate policies in the rural sector. Hence, the reality of the farm household is really that of a 'multisectoral firm', and traditional monosectoral policy approaches are inadequate (Commitment 2).

*    An agriculture based on knowledge and human capital that invests more in human capacity building and in the development of producers' roles as market agents and entrepreneurs (Commitment 3).

*    An agriculture responsive to the feminization of the countryside. Agricultural tasks often are concentrated in the hands of women. Women typically also work more hours than men and for less or no income (Commitment 1).

*    A sustainable agriculture in which the use of resources is integrated with its conservation, with new technological matrixes that correspond to its productive heterogeneity, supports peasant production and guarantees sustainable development (Commitment 6).

*    An urban agriculture. It has been estimated in 1994 (Mougeot) that about 12 percent of the world's total population - 700 million people - are supplied with food by 200 million urban farmers - agriculture is no longer an exclusively rural function (Commitment 7).

*     Globalized agriculture refers to the growing influence of globalization on national agricultural systems, with particular reference to increased linkages between agriculture and international finance and the growing influence of international competition. (Commitment 4).

2.These changes need to be borne in mind throughout the discussion below, as they provide some insight to the context in which processes of land reform operate.


3.Land tenure reform is usually part of a broader economic and political reform process. In order to reflect the temporal/historical dimension, three 'types' or 'generations' of land reform can be recognized. The first refers to land reform in which land is issued or redistributed by the state according to defined discretionary rules. The second refers to cases in which land is purchased for redistributive purposes, and the third refers to cases in which land reform occurs in the context of a comprehensive supporting institutional framework that enshrines rights and security. And, of equal importance, the third generation type of land reforms is distinct in that it does not concern itself solely with landless groups, but also seeks to utilize reforms as a means of strengthening the economic and productive potential of existing producers who are constrained by pre-existing tenure arrangements and institutional dysfunction. These considerations lead to the following conclusions:

(a)    Land tenure reform does not refer uniquely to a process of redistribution. It is also a means of acknowledging existing arrangements on the part of farmers who are already in possession of land.

(b)    Land tenure reform is basically an institutional reform and must therefore be accompanied by other reforms, especially in terms of policies and institutions. New investments will not be sponsored by land tenure reforms alone. Tenure reform is a necessary precondition but not an investment mechanism in itself.

4.Food security has three dimensions - availability, access and stability - and various levels of aggregation - global, national, household and individual. Given this multi-dimensionality, it becomesclear that the achievement of universal food security at the individual level, which implies similar achievement at all subsequent aggregate levels, is either constrained or facilitated by a combination of social, political and economic conditions.

5.Food in security and poverty issues should take into account the processes of transition from what have been state-led economies to more market-led economies. One clear constraint in the transition process, especially in agriculture, is the inadequacy of the necessary legal and institutional framework to support the functioning of competitive markets. Markets do not function in a vacuum, they require information, rules that govern the market actors, and the establishment of enforcement institutions.

6.Many of the food insecure countries depend heavily on agriculture in terms of percentage of rural population and/or contribution to GDP, and poverty continues to have its greatest incidence in the rural areas of developing countries with the landless and the near landless constituting the largest single group of those afflicted by poverty. Agricultural production and productivity is a high priority in achieving progress towards food security for all and in reducing rural poverty; land reform plays an important role in advancing towards both goals.

'Third generation' land reforms

7.The experiences of the first and second generation type land reforms are instructive. Even those that can be considered 'successful' in terms of providing access to land to previously denied groups, often failed to provide either appropriate policies or suitable institutional settings conducive to the promotion of sustainable rural development for the smallholders. In contrast, the third generation type of land reform focuses principally on institutional reforms, or more precisely, considers the role of agrarian institutions as key elements for the success of food security. Firstly, the third generation type reform considers access to productive assets - of which land and water are decisive - as the crucial condition for effective policy alleviation. Secondly, it takes into account the multiple sources of income at the rural household level. Thirdly, it focuses on creating the conditions underlying the inverse relation between total factor productivity and farm size based on the superior advantage of small farms in terms of costs of supervision due to the use of family labour. Fourthly, it recognizes that for these types of reforms to be successful the full participation and intervention of civil society organizations - from the local to the national level, from community groups, advocate NGOs, technical NGOs, farmers' associations, chambers of agriculture, etc. - are needed and should be deliberately sought. Fifthly, these reforms are based on alliances and coalitions. These coalitions are convergent, that is to say they are coalitions in which a diverse range of actors with different and even antagonistic views, coalesce around a number of limited issues from which they all stand to gain. Sixthly, although its main purpose is to combat rural poverty it recognizes different paths to do so.

8.These third generation reforms cannot prescribe linear or deterministic processes. Similarly, they must not be maximalist in the sense that they enshrine a particular goal - such as the eradication of poverty - without the attainment of which nothing else can be done. And finally, third generation land reforms recognize that the bottom line refers to production, productivity and competitiveness, in that they constitute the basic foundation of sustainable rural development. The next section discusses what has become another of such foundations.


9.The importance of off-farm employment and incomes can be looked at both from the point of view of household objectives and strategies, as well as of rural communities and rural development for food security. Income diversification can be a powerful instrument for households to de-link consumption from farm production and income fluctuations (consumption smoothing). The importance of off-farm income in terms of shares of off-farm income in the total income of farm or rural households, as derived from surveys carried out in three developing regions, is shown in Table 1 in the Annex. Although there is a large variation in the shares of off-farm income, the data reveal that in most cases such income is a very significant source of livelihood for rural households.

10.The use of off-farm income as an investment source constitutes an important link between the household and the rural community. The use of off-farm income for investment and improved productivity contributes to an overall improvement in economic activity in rural areas. This is particularly true for agricultural growth, which may result in generalized rural development throughthe creation of activities and employment opportunities up- and downstream of agricultural production, while increases in agricultural productivity will result in release of agricultural labour force for labour demands from elsewhere in the economy.

11.A key conclusion from analyzing the impacts of macro-policy through the lens of micro-economic decision-making models is that access to various forms of capital is pivotal in determining household income strategies and, therefore, the likely change in household behaviour and well-being when faced with macro-policy changes. Carter and Mesbah (1993) conducted a simulation analysis of how various strata of the Chilean peasantry fared in the land market during a period of rapid agro-export led growth. The key in this model and, hence, in defining the behaviour of various classes of rural households, is that households are stratified according to resource endowments (asset typologies) which imply different strategies. Each strata is differently constrained by its relative need for and ability to participate in size-sensitive rural factor markets.

Household activity choices under incomplete markets

12.In general, we can distinguish two sets of factors that affect the decision by households to diversify income sources. One set is usually associated with the risk motive including low and unstable yields, short growing seasons, lack of irrigation, credit/capital/insurance, market failures and land constraints, and associated mostly with risk-reducing motives of households. The other set includes a number of 'pull' factors that induce reallocation of resources towards non-farm activities to take advantage of income-increasing opportunities off-farm. They include inter alia intersectoral factor returns in favour of off-farm activities, migration opportunities and opportunities in sectors up- or downstream of a growing agricultural sector.

13.The income generating activities of farm and rural households depend on their control over and access to assets. A farm household's portfolio may include a wide array of assets: land and water, productive capital, savings and food stocks, human capital, organizational capital, social capital, investments in small off-farm businesses, and migration capital in the form of access to a migration network. The asset portfolio of households will determine to a large extent the farm household's allocation of resources among various combinations of farm and off-farm activities. Changes in the asset holdings of a household in favour of an asset are likely to cause a shift in its resource allocation among income generating activities. Income derived from off-farm activities will, in turn, influence the household's ability to acquire assets and thus the earning capacity of the household in the future. Households are forward looking entities which, under the particular set of constraints they face, use (part of) income to build an asset base that better corresponds to their needs for income and security. What one observes at one point in time (as for instance the correspondence between asset holdings and income earning sources) is the result of previous decisions and actions involving asset accumulation or a lack thereof.

14.Decisions made by the farm household concerning its set of activities thus depend on the incentives it faces as well as on its capacity to take advantage of such incentives. Households facing similar technologies and risks and with similar attitudes towards risk may invest resources differently if they are facing constraints due to their different income and asset profile and collateral capacities. Private investment in agriculture is thus determined by a variety of factors, such as tenure security.


15.The impact of titling and tenure security on credit availability and agricultural productivity can be broken down into supply and demand effects. Demand effects occur when the acquisition of land title increases the farmer's security and certainty that he/she will be able to maintain possession of the land and benefit from investments, improving the productive capacity of the farm. Increased security is hypothesized to enhance investment incentives and increase the demand for capital and variable inputs complementary to capital and, thereby, raise agricultural productivity. Supply effects result when provision of secure and legal land title improves a farmer's access to cheaper and longer-term institutional credit because land can be pledged as collateral for loans.

16.Land registration may reduce risks and transaction costs for some landowners but, particularly for small landholders who formerly depended on customary tenure mechanisms, risk and transaction costs may actually increase (e.g., in Kenya). With regard to credit access, financial markets in Africa often are not available for small farmers, even to farmers with titled land; bank credit going to agriculture may be determined by factors other than profitability. Land as collateral is less attractiveor useful in Africa as financial institutions are reluctant, because of cultural norms and administrative problems, to foreclose on landed property. Off-farm income is increasingly viewed as a better candidate to serve collateral requirements, particularly in areas that feature at least some (not necessarily formal) savings institutions.

17.Investment in land improvements and new technology has taken place in Africa whether ownership is formalized or not, and generally indigenous tenure systems have not really had a negative impact on productivity. While definitive and conclusive studies on the long-term effects of the introduction of land titling on agrarian structures still need to be undertaken, it appears that land titling alone may not be sufficient for the achievement of substantial improvements in levels of agricultural productivity. It may lead to more mobilization of rural savings (where these are available), but if more credit is not made available specifically to smallholder agriculture, credit transactions may not necessarily increase, and neither will land transactions. What is therefore needed is a broad institutional environment and legal framework conducive to longer time horizons for decisions taken within the sector. This may include factors such as the credibility of government, and whether or not there are other devices that can be used to formalize the goverment's commitment towards smallholders.

What Kind of Participation and Access do Land Markets offer the Rural Poor?

18.Land markets tend to be very local and informal in terms of documentation of transactions and property rights and in terms of information dissemination. Land is obviously not a homogeneous commodity and in many places its quality varies even within micro regions. Transaction costs are often very high creating barriers to entry for the poor and disincentives for participation on the part of holders of large tracts of land. In many circumstances, until recently, government rules restricted or prohibited various forms of transactions in land. In these circumstances, much informal activity continued but was largely unrecorded and often only use-rights were transacted. Finally, the nature of informality means very uneven access to information. For example, availability of land is often announced only through kinship and friendship networks.

19.In Central and Eastern Europe land markets did not exist are being developed now. This occurs in a context of confusion over allocation rules for privatization of state farms and a fear of further proliferation of tiny, fragmented farms. While the modus of land reform of the past needs to be structurally adjusted, the policies of structural adjustment of today need to encompass some forms of land reforms (e.g., land market development policies that explicitly address the ability of the rural poor to participate; issues of land tenure insecurity). In particular, two differences in present land policy are: (a) the use of market-oriented strategies and (b) the recognition of the interlinkage between land markets and other rural factor (capital and labour) markets.

20.Even if land markets are competitive, it is unlikely that land will shift to the landless and the land poor. In order to finance a land purchase in competitive capital markets, the poor have to dig into current consumption which they cannot afford to do. In this context, segmentation means that there are systematic forces limiting the fluidity of land transfers from larger to smaller farms. Segmentation is also reflected in the fact that land price per unit is systematically and significantly higher for small parcels than for large parcels; most sales occur between farms of similar sizes.


21.In Asia, especially South Asia, poor households tend to derive most of their income off-farm, which reflects the 'residual' nature of their employment; they may be willing to accept a lower income on average to attain lower variability of income streams. For middle-income households, which often specialize in land-intensive crop production, the off/on-farm income relationship is rather equal. Higher income households derive more income from off-farm sources, since they also have better access to education and are better able to diversify investment and employment to attain higher wages in productive employment.

22.The characteristics of agricultural production, especially land/labour relationships, constitute another determinant of access to off-farm incomes. All other things being equal, one expects that in areas with very low land-to-labour ratios (such as in most of Asia) expanding population and inadequate landholdings will 'push' poorer households out of agriculture and towards off-farm income. In Africa, on the contrary, abundant but often low quality land and scarce labour tend tokeep poor households in agriculture, with better-off households more able to respond to incentives for participation in off-farm activities.

23.If education is easily accessible by rich and poor alike, then off-farm income may have an equalising and poverty-reducing effect. Such an effect may be enhanced if technological innovations in agricultural production are such that smallholders can take advantage of them.

Effects on Productivity in Agriculture

24.The nature of asset holdings and types of constraints faced by households will largely determine both the motives and impacts of off-farm income. The distribution of assets by their degree of liquidity will be important in the household's decision to diversify income. Given access to credit, the incentive of households to diversify will be stronger the higher the share of illiquid assets in the asset portfolio. Similarly, the more illiquid the assets and the harder the access to credit, the bigger the impact of off-farm income is expected to be at the margin.

25.Off-farm income-earning opportunities may substantially affect agricultural productivity through changes in the willingness and ability of households to undertake risk. Increased income from off-farm employment or small business activity may enable a farm household to: (a) increase the area under cultivation; (b) use more purchased inputs due to both increased liquidity and increased security in case the crop fails; (c) diversify farming activity into new crops and livestock or expand the cultivation of cash crops. On the other hand, off-farm activities may compete directly with new technologies for farm household's labour resources preventing the adoption of labour intensive new technologies.

26.While off-farm incomes can be used as collateral for farm asset acquisition, it increases access by rural households to agricultural assets (including education and training) the 'virtuous circle' effect is obvious: agricultural growth opens opportunities to households for employment and income diversification, which in turn gives incentives for increased investment in agricultural assets and further growth of the agricultural and the rural sectors.


27.A review of issues on land tenure and off-farm income leads to the overall conclusion that a multiplicity of factors affect the incentives and capacities of households to diversify their income-earning activities and to achieve tenure security.

28.Land tenure policies therefore must be formulated in a much broader setting than heretofore, and approved at a higher level than a ministry, unless its mandate extends to the whole chain of food production and rural employment as well as land tenure.

29.Three major conclusions emerging from the above review highlight the following:

(i)        the importance of analyzing the relationship between policy formulation and institution-building, with special attention to the role of local institutions and local governments in the fields of agriculture and rural development;

(ii)    the recognition of the dynamics of income-generation strategies at household level and the related need for surveys, case studies and comparative analysis for better-focused policy interventions;

(iii)    the need to promote partnerships and alliances between farmer associations, private sector, academic or research institutions and governments, as a strategic choice to include stakeholders, where the complexity of the food production chain is better understood.

30.In view of the crucial importance of these issues for FAO, COAG is invited to provide the necessary guidance for required action by FAO and governments.


Table 1

Percentage share of non-farm earnings in total rural cash income

in selected countries in Africa, and in total rural income

in selected countries in Asia and Latin America (1)






Liberia, Western Region*






China, Yunnan Province









Mexico, Ejidos



*    Includes all rural households, not necessarily all farm households. Sources Liberia, ILO (1982a); Tanzania, ILO (1982b); China, FAO Investment Centre (1992); Pakistan, Adams (1994); Chile, Valdes (1995); Mexico, de Janvry et al. (forthcoming).


(1) The definition of farm income is not uniform across surveys. For instance, in some surveys only crop income is considered as farm income while in others livestock income is also included in the definition.