Michael Lipton (Sussex Univ.) & Yashar Saghai (Johns Hopkins Berman Inst. Bioethics) 9/9/2014
The focus of this paper is on ethical issues surrounding farmland redistribution through land reform to increase food security. As the main agent of land reform is the state, we explore the overlapping territories of moral and political philosophy. The ambition of this paper is two-fold. Our first goal is to provide an overview of evidence about facts and consequences of land distribution and redistribution and to compare it to feasible alternatives. Our conclusion is that private land redistribution (but emphatically not collectivization) in low-income and most middle-income countries with very unequal land control has not only raised poor people’s share of income, but also accelerated, or at worst not decelerated, growth. It has reduced poverty and inequality, and increased equality of opportunity and food security. Our second goal is to clarify some crucial issues in farmland access ethics by exploring the roots of the moral case for and against land reform: the competing claims of equality of opportunity and incumbent legitimacy based on property rights. The received view is that this fundamental moral disagreement on land reform is intractable. We disagree. We identify some morally relevant considerations that, to the extent they apply to the case at hand, alter the weight assigned to equality of opportunity or legitimate incumbency claims. We argue that advocates of these competing claims can endorse, minimally converge on, or at least not reject this approach that is responsive to their claims. If we are right, it is then possible to deliberate on the moral permissibility and desirability of land reform on a case-by-case basis.
Why should a workshop on global food security and ethics consider land access, distribution, and reform? From a food ethics perspective, avoidable mass undernutrition is undesirable, arguably impermissible. Fortunately, avoidance has proved feasible. The proportion of underweight children in developing countries declined from 25% to 15% in 1990-2012. The UN Millennium Development Goal of "halving hunger" in 1990-2015 has been achieved at regional level in Northern Africa, East and Central Asia, and Latin America (WHO 2014; UN 2014).2
Moreover, from many local and national successes, we know how to slash undernutrition: through either increased food production and/or more pro-poor distribution (the latter, incidentally, would also reduce overnutrition).3 Yet in South Asia and sub-Saharan Africa, 25-50% of people (and 30-60% of under-fives)4 remain undernourished.5 As a result, many die or suffer lifelong impairments.
Over 80% of undernourished people live in rural areas. Most depend on agriculture, mainly as small farmers and/or farm labourers cultivating food staples. The incidence and severity of undernutrition among agriculturists are almost everywhere much higher than among other groups (Lipton 2013). Thus most people who cannot obtain sufficient food live from growing it.
Their plight is due to a plurality of factors, which classify into three groups: those affecting: (1) scarcity of farmland-per-person, (2) low conversion rates of land into basic human nutrients, and (3) in many countries very unequal distribution of land and income.6 The moral imperative to reduce or eliminate mass undernutrition can in principle be met by attacking (1), (2) or (3). Let us take each in turn.
As for the first issue, farmland scarcity (even in Africa) continues to sharpen, for two reasons. Especially among poor rural people, population continues to grow; and population aged between 15 and 65 —the main group seeking work, including farmwork, to obtain income and thus food—is growing faster (e.g., at 1.5% per year and 3.0% per year respectively in Bangladesh). Also a very high proportion of arable land is already farmed; what remains is less fertile and/or costlier to bring into cultivation; and land is being lost to soil and water depletion, or converted for buildings.
As for the second issue, low conversion rates of land into basic human nutrients, crop yield rises due to the ‘green revolution’ have slowed; land continues to be diverted from human to animal food, buildings, and depleted waste. The conclusion is that, while some action on the scarcity of farmland and low conversion rates is feasible, morally permissible7 and on-going,8 more equal distribution of land and income might be a key policy option against undernutrition. More specifically, state-led land reform is one important redistributive tool for addressing undernutrition.
Land reform comprises laws with the main goal of reducing poverty and inequality by substantially raising the proportion of farmland controlled by the poor, and thereby their income, power or status. It raises intriguing and significant ethical issues only if it has been proved feasible, and if it is not, as many claim, ‘dead.’
Over 1.5 billion agriculturists today are affected by post-1950 land reforms, so it was feasible. Our question is: under what conditions (if any) is such state-induced land reform—from a moral and political standpoint—mandatory, permissible, or unacceptable? What of ‘counter-reform’: private, state-permitted or state-executed land grab from small farmers? (On land grabs, see Madison Powers’s contribution to this volume (2014)).
In this paper, we show that food ethics requires attention to the related area of farmland access ethics. Food ethics can be defined as the action-guiding inquiry into the obligations and responsibilities of agents (individuals, groups, corporations, and states) and the rightness or wrongness, as well as relative desirability, of policies and laws that affect the production, distribution and consumption of food. Farmland access ethics is the moral evaluation of actions, policies and laws that affect farmland distribution, allocation and use.9
Our focus is on ethical issues surrounding farmland redistribution through land reform. As the main agent of land reform is the state, we explore the overlapping territories of moral and political philosophy. Our main concern is with the obligations and responsibilities of the state, as well as the limits of state action, with respect to the interests and rights of its citizens and residents.
We hope this paper and the “Feeding the World, Ethically” meeting can help to sort out normative and factual assumptions behind conflicting positions with respect to land reform as a tool to address undernutrition. Land reform is a controversial area where ideological commitments often replace ethical argumentation and careful examination of the best evidence on its consequences. Perhaps surprisingly, the positive evaluation of, at least, outcomes for local production from certain types of land reform—specifically, the production outcomes of not-too-unequal smallholdings in developing countries—is largely consensual among agricultural economists, while it is little known to, and/or controversial among, other economists, including macro-economists and the policy-makers they advise. The evidence for and against land reform is rarely examined in the literature on the ethics of land access, which focuses almost exclusively on principled, rights-based objections to land reform.10 Our goal here is to show that the ethical status of land reform as a tool to increase food security requires a careful examination of the consequences of various types of land reform versus feasible alternatives and a case-by-case moral evaluation of the acceptability of land reform, despite right-based objections to land reform.
To do this, we first briefly discuss land reform, its history, and the ethical goals it is designed to pursue, which are primarily reducing poverty and inequality of opportunity and outcome. Of course, we acknowledge that other widely shared policy goals, seldom specific to land reform, matter too: liberty, efficiency, security against severe instability, and sustainability.
Next, we examine the evidence about the consequences of land reform, because they are relevant to the ethical evaluation of land reform with a specific focus on its contribution to food security. This includes not only the extent to which existing evidence supports the claim that land reform decreases poverty and inequality (its primary goals) and increases equality of opportunity and liberty for the rural poor, but also the impact of land reform on the economic interests and general well-being of others, such as the urban poor, landowners, and society more generally. We then explain why less heavy-handed alternatives to state-led land reform as a way of getting farmland to the poor may well not achieve these goals. Our particular focus is land reform’s most serious competitor, reform of tenancy regulation.11 We finally review the main principled ethical objection to land reform: that the rights-based considerations favouring the legitimacy of the incumbent landowner override the moral considerations favouring land reform based on equal opportunity. The received view is that the fundamental moral disagreement on land reform is intractable. We disagree. We argue that it is indeed possible to deliberate over the moral permissibility and desirability of land reform on a case-by-case basis. We identify some morally relevant considerations that, if present, increase or decrease the weight of claims for or against land reform.
What Land Reform is and Why it Matters Morally12
The aim of this section is to provide the reader with an overview of land reform and the debate surrounding different types of land reform, its history, and the ethical goals it is designed to pursue.
‘Classic’ land reform sets limits (ceilings) to land areas, allowing for quality, that one person or household may legally own. The vesting authority obtains above-ceiling land (with total, partial or no compensation) and then (1) farms it (state farming), (2) distributes it to co-operatives or collectives (co-operative/collective farming), or (3) redistributes it as individual smallholdings (smallholder farming), ostensibly to the farming poor. As a rule, (1) and (2) have performed very badly. But (3) is a different story, though often tarred with the same brush by its opponents.
Why has land reform followed by state or co-operative/collective farming performed badly? That is partly due to state violence, extraction, remoteness or error in the process of land redistribution and subsequent farm management, but more fundamentally because farming is ill-suited to joint production or distant management. Farmwork is spread out over space and takes much time to produce crops. Hence neither co-operative/collective farmworkers, nor state farm managers, can readily observe specific actions of a specific co-worker, let alone their consequences.
In contrast, land reform followed by redistribution to individual or family smallholdings, has proved much more successful, mainly because in low-income developing countries farming is well suited to holdings. For these, labour screening, search, instruction and supervision are relatively simple, but costs of borrowing and capital management are higher. So smallholders use more labour and less capital per hectare than largeholders. Where labour supply is ample and fast-growing, but capital is scarce—as in most developing areas - smallholders therefore cultivate each hectare more intensively, making better use of resources, raising income for labourers as well as for themselves, and producing more per hectare-year.13 What Happened
Today, post-1945 land reform provides substantial extra income from land and farmwork to perhaps half a billion people. Those reforms have reduced land inequality by distributing private rights from large owners to small and landless agriculturists. Such reforms affected, in succession, Japan, East Asia, much of South Asia and Latin America, and some of Africa.
That said, we acknowledge that between 1910 and 1980 even more land reform has involved a ‘terrible detour,’ away from very unequal landholdings, but towards state farms or co-operative/collective farms. State or collective farming was imposed, in succession, in Mexico, the USSR, Eastern Europe, China and parts of other Asian, African and Latin American countries. This hardly ever achieved the poverty-reducing goal of land reform that we discuss below; as for the equality goal, wealth, power and status often passed from an elite of rich farmers not to the poor, but to a new elite of collective and state farm managers. Further, big state and collective farms proved excellent vehicles for forced extraction of underpriced food, fibres and fuel (timber), helping the urban poor (and rich) but harming—and disincentivising—the more numerous, poorer and weaker rural poor.
When these types of land reform failed, decollectivisation followed—starting with the spectacularly successful ‘household responsibility system’ in China in 1977-84. There, in Vietnam, and in many other cases - but not all, and not in most of the former USSR—decollectivisation led to small, not-very-unequal farms. Such land reform by detour now affects over a billion people dependent on agriculture.
Today, land reform is often pronounced dead. But the diagnosis is mistaken because it conflates the fate of land reform leading to disastrous state and co-operative/collective farming with that of land reform strengthening smallholder farming. Thus, the prospect of land reform is still alive, for example in Bolivia, Brazil, China, the Philippines, South Africa, much of the former USSR, Venezuela, Vietnam, and Zimbabwe.14 It is widely discussed in cyberspace. Social movements like La Via Campesina, and pressure-groups/think-tanks like the International Land Coalition, provide a global frame for further smallholder-friendly land reforms (though some food sovereignty activists seem to waver between smallholder farming, co-operative farming, and outright collectivisation (see Agarwal 2014)). Granted, in rich countries with ‘scarce’ rural labour but relatively plentiful capital, land reform is unlikely to play a major role in policies aimed at reducing poverty, inequality, or food insecurity, especially since agriculture there seldom engages above 5-10% of workforce (much less in NW Europe and the USA). However, land reform of the right type matters in many developing countries where underemployed labour is high, capital is scarce, and the majority of the population lives in rural areas. As we shall see, the great weight of evidence is in favour of redistributing land as individual smallholding through certain types of public land reform because smaller, less unequal farms usually conduce to social efficiency and other positive outcomes, including equity.
To sum up, directly or by detour, land reform has increased the farmland and work prospects of at least 1.5 billion people and has raised their power, status, and income—especially where amplified by a ‘green revolution’ of irrigation, fertilizers and new seeds. They are less poor, or not poor, as a result, and have therefore access (through production or purchase) to more food and/or food of better quality.15 Goals of Proponents of Land Reform
Reducing Poverty and Inequality of Outcomes
The case for state action to reduce inequality of outcomes16 - and even to reduce poverty - remains contested, for a variety of reasons that we will not rehearse here.17 We simply note that the case is stronger where poverty is severe, widespread, and a major cause of undernutrition, illness and death, as in most low-income countries; and where poverty and affluence (and hence inequality) are ascribed, e.g., by inheritance, rather than achieved by work, skill or saving. Achievement partly reflects differences in effort, willingness to defer gratification, and success in offering what is in demand. Hence some achieved inequality rewards these desirable features of individuals and helps economic growth. In contrast, ascribed inequality largely reflects luck; asset inequality is more clearly growth-inhibiting than income inequality.18 The more of a country’s income accrues by ascription, including inheritance, the less is available as income to reward or to incentivise those who meet consumer demand through their work, skill or enterprise. A part of assets is inherited, and income from such assets is clearly ascribed rather than achieved.
The case for redistribution in developing countries applies with greater force to land, which is particularly likely to be ascribed rather than achieved. Inheritance and inter-family marriage are the main means of accumulating land, as in Middle Eastern countries (El-Ghonemy 2003). Reducing farmland inequality is also especially likely to be effective against poverty, for several reasons. First, it addresses vulnerability, a main cause (and curse) of rural poverty: extreme land inequality leaves the poor and near-poor with little or nothing to mortgage or sell in bad years, so they are vulnerable to slipping into deeper poverty. Second, land—unlike financial assets—can be combined with the poor’s main resource, labour, helping them work their way out of poverty. Third, probably over two-thirds of the world’s dollar-poor have some experience of how to manage farmland.
Equality of Opportunity
It is useful to distinguish inequality of opportunity from inequality of outcome. However, some forms of inequality of opportunity imply inequality of outcome. This is obviously the case where inheritances are unequal. Even in one generation, very unequal property ownership denies equal opportunity to those who own none or little. Equal opportunity to afford a Porsche matters little. Equal opportunity to obtain sufficient food matters a lot; in principle it can be guaranteed even in societies with very unequal property, but the very poor seldom have the political power to achieve this, nor the very rich the will. Also, very unequal property means that big property-owners (hereafter called ‘owners’) have much better chances than others (‘non-owners’) to obtain high-quality education and health care, favourable work and leisure environments, and chances to turn these into a long and happy life—all, by the way, correlated with better nutrition. Further, very unequal property, by yielding property income to owners, also implicitly taxes opportunities for, and incentives to, non-owners. Owners’ property income absorbs GDP that might otherwise reward people—including non-owners—who work harder, manage better, or are more able.
The damage to non-owners’ access to equal opportunity is, as a matter of social fact, more widely tolerated if property ownership is achieved mainly by saving, skill, effort or other achievement, rather than by inheritance or other ascription. The damage to non-owners’ access to equal opportunity is more severe, therefore, for types of property such as land that are more frequently inherited; and for forms of property, such as farmland in mainly agrarian societies, that form very large parts of the asset base, especially of the local asset base, conveying power to grant or deny employment or credit to non-owners. Most people agree that a just property distribution should not be so unequal as to cause major, avoidable harm to non-owners’ access to equal opportunity.
The main goal of land reformers in enhancing poor rural people’s access to land is to reduce poverty and low-end inequality, not to increase liberty. Indeed, the main reasoned ethical argument againstreform is that, by reducing property rights, it reduces liberty. Yet many land reformers justify land ceilings—which focus on reducing top-end land inequality—by arguing that the distribution of such land to the (near)landless poor increase liberty, that is, give them some measure of control over the shape of their lives by enabling to access key options that are entry points for other options and serve as barriers to domination by others.19 Having land of their own, the poor rely less for work, rentals, emergency loans, or trade on one or two local ‘rural tyrants’ (Bell 1990, 143-66)—almost always major land controllers, but often also employers, landlords, lenders, traders, with interlocking market power over things that the local poor can neither live without nor, in many cases, readily get elsewhere.
In a nutshell, we have argued that the public debate about the merits of land reform is biased against land reform because it confuses different types of land reform. While land reform followed by state farming or co-operative/collective farming has by and large failed, sometimes dramatically, the same conclusion does not hold for redistributive land reform that promotes smallholder farming. The latter type of land reform has been widespread and rather successful, and remains a genuine policy option for some developing countries. Proponents of land reform usually justify their position by appeal to the importance of securing equality of opportunity, though increasing liberty and other goals might also count in favour of redistributive land reform.
The Consequences of Land Reform
After this overview of land reform, its history, and moral goals, we need to look critically at the consequences of redistributive land reform and ask:
What is the evidence that the consequences of land reform are as advertised (less rural poverty and inequality, thus more equality of opportunity, more access to food, etc.)?
What are the consequences of land reform for groups and persons other than the rural poor and for society (generally)?
If consequences referred to in these two questions are largely positive, the moral case for land reform is bolstered; if these consequences are mixed or largely negative, the case for land reform becomes, from an ethical perspective, more complicated.
Ethical controversies about farmland access may be softened if several conditions are met. First, one must endorse the view that ethical analysis needs to take into account consequences20 in judging the rightness of state actions; second, those consequences must be in fact overall positive. The area of plausible ethical disagreement is then narrowed by new knowledge about facts and consequences. Our claim is that a wealth of accumulating evidence shows that redistributive land reform has favourable direct or indirect (via its impact on poverty and inequality) consequences on access to food:21
Widespread caloric and micronutrient deficiency is widespread, but normally diminishes quickly with rapid growth and poverty reduction (except in a few important cases, e.g., north India).22 It is also harmful to productivity as well as to lifelong health, well-being, flourishing and choice. Of course, various sorts of public action (regarding health, agriculture and land) reduce undernutrition, in particular during childhood. But the evidence shows that land reform helps: control of farmland, even in small amounts, substantially raises intake of calories and vitamin A.
Information on the size-distribution of farms has become more readily available. In almost all the low and middle-income countries with more than one of the (in principle ten-yearly) ‘rounds’ of the FAO World Census of Agriculture, land is passing from larger to smaller operated farms, confirming the theory and evidence for the efficiency of small scale. But, where there is little land reform, the process is very slow. Farmland per person, and per rural worker, continues to fall in East, South and Central Asia, and parts of Africa, due to growing rural populations, even faster-growing rural numbers of working age (despite urbanization), and loss/unsustainability of farmland.
Close connections have been established between (a) poverty and associated caloric and micronutrient undernutrition, and (b) rural residence, unequal and for the most part tiny landholdings, and landlessness. Since the 1960s, sharp retreats of mass poverty have been observed in the wake of redistributive (but not of collectivising) land reform, especially alongside agro-technical progress and reduced population growth. This has been seen in many countries and political systems, but perhaps most strikingly in China after the land reforms of 1977-1984 (as mentioned in the previous section).
Great progress in reducing poverty and undernutrition has been made in most of Asia and Latin America. Such progress has, arguably, almost always been initiated by greater policy attention to, and spending to improve, land-water productivity in smallholder agriculture—usually accompanied either by initially not-too-unequal smallholdings or by some farmland redistribution.
Contrary to what is often assumed, large farms are not always more efficient and innovative than small farms. Indeed, a large majority of agricultural economists is persuaded by accumulating evidence that in low-income, mass-poverty environments not-too-unequal, smallish, usually family farms—mainly but not solely because their low unit transaction costs of labour supervision—stimulate more labour-use per hectare, and therefore produce more per hectare-year, than do large unequal farms and farming systems (Eastwood et al. (2009); Hazell and Rahman (2014)); Lipton (2009, chapter 2)). This empirical finding is crucial to the force of the ethical argument for land reform on grounds of equal opportunity.23 That argument would fail if reform actually reduced farm output, especially of food, and that reduction were sufficient to outweigh any gains to the poor from redistribution. Fortunately, the data show that small farms normally produce more output, and food, per hectare than big ones. However, one could ask whether land redistribution, at least in the short run, might disrupt traditionally large-farm services such as credit, irrigation, fertilizer provision in small packages, marketing, and input provision, or even deny to small-farm reform beneficiaries these services. These are problems that cannot be overlooked. But there is also much evidence that, with sensible policy, even very small farmers can be well served by markets, by (genuine small) service co-operatives, and/or by temporary state subsidy or provision.
How do farmland access regimes affect sustainability of natural resource use? Again but differently, accumulating evidence may ‘blow away’ some ethical disagreements, provided we reach some level of agreement on what sustainability means and how to measure it. The evidence here is that smaller farm size and more equal land distribution improves some aspects of sustainability (e.g., those calling for labour input in slack seasons) but worsens others (e.g., those calling for collective agreement to joint action). For instance, small farmers are likelier to use compost or mulch, but less likely to jointly manage water effectively down a shared watercourse. If sustainability should be served—if the rights and benefits of future populations ought to loom large—then different land-rights regimes and distributions require different arrangements: by farmers themselves, market providers, or the state as regulator, provider, subsidiser, or producer. However, the evidence does not suggest that small and/or equal, as such, is either more or less sustainable than large and/or unequal (Lipton 2009, 45-8).24
As the most important issue is the third, we return to it in more detail. Will land reform into smaller, more equal farms increase output and income as well as the share of the poor? The poor might lose if reform reduced output; the EO case for reform would then be destroyed. Happily, the great weight of micro-evidence from all over the world25 shows that in low-income labour-surplus economies:
More productive use of land and capital, higher levels of labour use and/or earnings-per-hour, and (hence and otherwise) faster and more poverty-reducing economic growth, occur if land control is distributed mainly into smallish, not-too-unequal private (usually ‘family’) farms,26 rather than concentrated in either a few large private farms, or in state or collective farms;
Such poverty-reducing27 growth is usually advanced28 by appropriate state-led redistributive (not simply tenurial) land reform, and (on weaker evidence) by appropriate state conduct on land grab and contract farming (other problems with these strategies left aside);
Such redistributive land reform since 1950 has been huge in scale and scope, land grab less so (Cotula et al. (2014), and contract farming much less so;
On-farm, offsetting small/family farmers’ lower unit labour-transaction cost (e.g., to supervise), they have higher unit capital-transaction cost (e.g., to borrow, or to use machinery). That looms larger as a country develops, accumulates capital, and urbanizes labour. So eventually (as in Western Europe, North America, Australasia, Japan) small-farm advantage gives way to large-farm advantage;
Off-farm, small farmers have special problems of collective action in water-system management, and in some post-harvest operations in scale and bulking-up, especially relating to supermarkets and export processors;
Such problems, while in principle offsetting small farmers’ on-farm advantages in low-income labour-surplus economies, are often effectively met by market provision, farmer co-operation—exemplifying the many areas where collective action by small private farmers has worked (Ostrom 1990)—or even state support.
Hence, accelerated smallholder-based farm growth has pioneered Asian and other ‘miracles’ of growth, poverty reduction, and subsequent urbanization and industrialization. It has often been enabled by redistributive land reform, and combined with public policy supporting agro-technical progress and farm-water control.
The route from land reform to these ethically desirable outcomes is not smooth. Evasion, avoidance, and backtracking on redistributive land reforms have been common, though less so than is widely believed—and with less damage if we allow for the effects of reform legislation on private land sales and inheritances, especially as the near-landless become aware of their new legal rights. Regrettably, gender equality has seldom been salient in most land reforms, especially in implementation; but the letter of the law, and even the practice, have often improved somewhat on pre-reform land law and custom. All in all, there is no doubt that in many countries land reform has greatly equalised farm size, and successive censuses of agriculture tell this story clearly.
Alternatives to Land Reform
In the previous section, we argued that the evidence for the positive consequences of land reform on food security is strong. But is state-led land reform the best way to attack poverty and inequality, and hence undernutrition—or even to redistribute farmland? The response is yes because alternatives to state-led initiatives are limited:
Religious and charitable bodies have historically dominated private efforts to reduce poverty, and in some areas still do, but usually concentrate benefits on the devout and accessible. They rarely seek to redress land inequality, in many developing countries the main source of total inequality (and arguably poverty). Indeed, historically religious foundations have been among the largest landowners.
Occasionally, ethics-based movements induce the rich to make significant transfers to the poor. A case in point is Indian reformist Vinoba Bhave’s land-grant Bhoodan movement in the 1950s. Another example is Brazil’s Movimento dos Trabalhadores Sem Terra (MST), a pressure-group poised between ethics-based advocacy and direct action by the poor.
Another option is, indeed, such direct action by the poor to redress poverty and inequality, ignoring, by-passing or overthrowing the state. Some have argued that such forms of redress, especially of land inequality, are the bedrock of future democratic polities (Moore 1966). However, few would prefer that poverty and inequality were corrected not by legislative process through a legitimate state, but by land invasions, revolution or civil war (including class war). Those violent alternatives have faced states run by groups so self-interested, incompetent, unaccountable or resourceless that they will not or cannot seek to reduce poverty or inequality, by land reform or otherwise.
A final option is voluntary private contracts, that is, tenancy. In rural Asia and Latin America, farmland tenancy remains widespread, and usually shifts farm management (and income from it) from larger, richer agriculturists to smaller, poorer ones.
Let us dwell on the scope and ethics of private tenancy and state-led tenancy regulation. Even with land ownership in large units, can some of the benefits for output, employment and poverty reduction, associated with smaller operated farms in low-income, labour-surplus developing countries, be achieved by tenancy? There is some evidence in favour of tenancy. Normally:
Tenancy transfers land, and income from its management, from richer to poorer rural people;
By making farms smaller, tenancy increases employment (and labour income) per hectare;
Tenancy can combine small farmers’ on-farm advantages with off-farm scale economies in processing and marketing: landlords - especially in sharecropping contracts - can find it pays them to sell marketing, credit, extension, bulking-up/transport, and other services to tenants.
Yet tenancy is unusual and/or highly controversial in many areas, often for good reason. It often involves one, or a few, big landlords with local monopoly power over many small tenants. That often gives big landlords power in other markets too: for labour, credit, and/or farm inputs or outputs. Tenants often have little security, and therefore (reputation apart) few inducements to invest or conserve land-water resources. Yet the proven efficiency advantages of small equal operated farms argue against state actions that restrict, and therefore discourage, tenancy where land redistribution is infeasible and land ownership very unequal. In such cases, landowners often evade or avoid tenancy regulation by open, implicit or tacit evictions, and then resume personal cultivation of the land. This defeats the aim of such legislation to increase control of land and hence income for small farmers and the poor.
However, tenancy regulation is much less at risk of being counter-productive if complemented by consensual land-ownership redistribution (with ownership ceilings) enforceable by a legitimate (e.g., democratic) state. Examples include Japan, Taiwan and South Korea in 1945-63, and to some extent elsewhere in Southeast Asia and parts of India and Latin America at various periods from 1945 to date. Nevertheless, even if successful, state actions relying on tenancy alone, as a means to reducing inequality of operated farm size, face difficulties. Though tenancy normally shifts farming from big to small units, increasing on-farm social efficiency in labour-surplus low-income economies, the contributions to poverty reduction are more modest. With land owned by self-operating large farmers, the poor get only the product of their labour. With land still owned by large landowners but rented to poorer tenants, the latter also get to keep some of the income from farm enterprise. However, the income from land remains with the owner. Also, to share the risk of farming, tenants and landlords in low-income environments, more often than not, settle on sharecropping contracts. Evidence suggests that any efficiency losses from sharecropping are small, but obviously, in return for bearing risk, the landlord can appropriate part of the tenant’s income from all sources—labour as well as farm management and enterprise. It is possible that tenants locked into such situations are no less poor than landless labourers, perhaps even poorer.
In sum, we have reviewed four alternatives to land reform, namely, reliance on religious or charitable organizations, ethics-based movements, direct action by the poor, and tenancy. We have argued that none is an adequate substitute for land reform.
Yet, one might reply that even if state action is indispensable, the state could avoid the use of compulsory means typical of land reform. Can the state achieve the same ends through non-compulsory means? Probably not:
Tax incentives to private charity may have a place in public policy, but come at the cost of other state actions.
Where formal activity and employment, in regular and accountable contact with the state, prevail—i.e., in developed countries—the main constraints on inequality and poverty are progressive tax; public spending on health and education; pensions; and benefits (‘welfare payments’). But this formula, typical of the modern welfare state, has not prevented an explosion of within–country inequality (though not of extreme poverty) in 1980-2014 (Piketty 2014). In addition, tax dodging and ageing populations threaten the feasibility and desirability of the formula. In our view, these problems are remediable and ought to be remedied. Nonetheless, and despite some partial successes of the modern welfare state formula in developing countries,29 this formula is much less relevant to the usually smaller and weaker states of low-income, substantially informal developing countries, and does very little to redress the underlying source of their inequalities of status and power: ownership of farmland.
A Principled Moral Objection Against Land Reform
We have so far focused on arguments for and against land reform based on the evaluation of its consequences compared to feasible alternatives. However, the argument for land reform to reduce gross inequality and poverty faces a fundamental, rights-based, challenge. Its underlying claim for equality of opportunity meets a counter-claim for legitimacy. When, if ever, is it efficient and just for the state to override property rights? Are rights in (farm)land somehow different? Most people agree that some property rights are historically legitimate, and that, for moral as well as economic reasons, property distribution should not be so disrupted as to cause major, avoidable harm to property right holders. Land reformers’ main goal, to reduce poverty and inequality, is usually based on the claim that grossly unequal land grossly violates EO. Their opponents usually retort that land reform disregards historical legitimacy.
If we accept this traditional way of framing the moral disagreement over land reform (opposition between two moral and political visions), this problem seems intractable. The challenge is to reframe the dispute in a way that helps to resolve it, or at least eases the tension between the demands of equal opportunity (hereafter, EO) and the historical legitimacy of the incumbent land owner (hereafter, legitimate incumbency or LI) in any particular case. This is the aim of this section. To do this, we first need to unpack the case for LI. Next we show that a context-sensitive evaluation of the moral permissibility and desirability of land reform is plausible, despite rights-based objection against land reform.
In practice, honouring LI means that incumbents can peacefully exercise, and legally enforce, legally acquired rights to property. Only then can incumbents in property, including land, meet their own legitimate expectations, and fulfil contracts. If property incumbents lack predictable rights, the risks of business are such that investment will be minimal, and production will decline and may seize up, with costs to everyone, including the poor.
The fight between EO and LI is the main conflict around the claim that land reform is a just route to the goals of reduced poverty and inequality. Notice, however, that there is an asymmetry: on the one hand, arguments for land reform sometimes include not only EO, but also claims to redress past violations of beneficiaries’ LI (e.g., in post-apartheid South Africa); on the other hand, arguments against land reform centre on LI, and can hardly ever credibly claim that land loss threatens big landowners’ EO. In poor agricultural areas, major improvements in EO normally require raising the status of the rural poor—the intended land reform beneficiaries—relative to big owners of farmland. But historical legitimacy often does not require freezing the claims of incumbents such as big farmers.
Indeed, land reforms based on restitution (as in South Africa) often do little more than nod towards EO, instead being based mainly on the claim that non-incumbents—usually persons of an assumed indigenous group—have legitimacy and the present incumbents do not. We shall return to these issues, but concentrate first on the main moral conflict between pro-reform and anti-reform views of land ownership, EO and LI. Grossly unequal rural rights to farmland are inconsistent with ‘labour equity.’ In crudest form, labour equity mandates that identical effort merits identical reward. In practice, modifications are needed for efficiency, with greater rewards where effort is more careful or skilled, involves management or entrepreneurial risk-taking, or is scarce relative to supply, and produces commodities that are similarly scarce. Labour equity, whether or not so modified, is consistent with a good deal of property accumulation. Some people may quickly consume all the rewards of their labour, while others use some of them to buy property. However, labour equity (and equality of opportunity) are not consistent with diversion of your labour income to me because I—let alone my grandfather—acquired all the land on which you might work, whether as slave, serf, tenant or employee. The inconsistency is more extreme if farming is a main source of income (as it is for the unskilled); and most extreme if I, alone or in a small group, can exercise monopoly power over local farmland, so you either work for (or rent from) me, or go without.
In the words of John Stuart Mill ( 2008, book 2, chapter 2, 40-41):
No man made the land. It is the original inheritance of the whole species. Its appropriation is wholly a question of general expediency. When private property in land is not expedient, it is unjust. It is no hardship to any one to be excluded from what others have produced: they were not bound to produce it for his use, and he loses nothing by not sharing in what otherwise would not have existed at all. But it is some hardship to be born and to find all nature’s gifts previously engrossed, and no place left for the newcomer.
Justice, however, does not directly imply that the state should compulsorily redistribute extremely unequal farmland. Justice is about not only equality (however conceived) or labour equity, but also legitimacy (and, arguably, about ‘rights’ as a whole). For the state to expel people from ‘their’ property seems arbitrary, open to dishonesty (blackmail, bribery), and, even if done honestly, open to terrible abuse, as the history of ‘ethnic cleansing’ illustrates. Distributive justice requires that the state should not, without strong grounds, deny LI. But why is any incumbent’s claim, even if historically founded, ‘legitimate’? Land reform, sometimes by mandated democratic governments, has often violated expectations that any claim to own and inherit land, once accepted in law, shall not be abrogated by state fiat. Yet LI claims remain widespread and durable. Based on them, owners big and small, including large hereditary landlords, make contracts, often with labourers, buyers, sellers, creditors, charities or poor relations. These, not just landowners, lose if those contracts must be broken due to a land reform. Yet EO, or at least absence of grossly unequal opportunity due to ascription, seems as central to justice (and long-term economic efficiency) as LI. What if land reform in low/middle-income agriculture-based economies is mandated by EO yet ruled out by LI?
Here, a traditional approach to the ethical debate on land reform fails to offer a formulation of the problem that is likely to ease the tension between conflicting views about the moral foundations of land reform. For some extreme libertarians, LI (and land stasis) almost always trumps EO (and redistributive land reform); the reverse is true of some people extremely confident in progress through purposive state action. In practice, no state—and no land policy—wholly driven by LI to the neglect of EO, or the reverse, could subsist, let alone be just and efficient.
Are there alternative ways of framing the problem? In other words, can we ask whether land reform is justified or unjustified without denying the claims of EO or LI? For instance, we may ask whether a just and efficient state should seek to precisely ponder, or maximize, some weighted sum of LI and EO. Such an approach, and an effort to ground the weighting in some notion of ‘happiness’ or wellbeing, is attractive to consequentialists (especially Utilitarians). But it faces many theoretical and practical challenges. How could any particular weighting be justified and implemented? Weights would differ among groups according to moral stance, self-interest, and perception of power,30 and would change over time. More fundamentally, the algorithmic weighting of LI and EO seems problematic because these are apparently ‘incommensurable’ goods or values. Some maintain that in the absence of a common currency to which good or values could be converted, they cannot be compared.31
Our thesis is that this way of formulating the problem is based on a mistaken view that incommensurable goods or values are also incomparable. But “just because [some values and goods] can’t be compared or ranked in terms of one master-value or formula, [does not mean that they] can’t be compared or deliberated between at all” (Chermiss and Hardy 2006). Indeed, they can be compared but only on a case-by-case basis, without any overarching rank-ordering of EO and LI or weighing based on assigning them a fixed value converted in a common currency (e.g., utility, whether or not based on willingness or ability to pay). Moral deliberation needs to be responsive to a number of considerations that are morally relevant in determining whether claims based on EO or LI have more weight in a particular context.
If our preferred understanding of incommensurability is correct, then the moral disagreement on the merits of EO and LI can be eased because context-sensitive moral deliberation is theoretically justified rather than ruled out. Moreover, this approach to deliberation is responsive to the claims of both sides. Stakeholders from both camps are therefore endorse this approach, minimally converge on it, or at least not reject it. Finally, this approach is preferable to more algorithmic views of moral deliberation because it accurately reflects the concrete use of ‘case-by-case’ arguments about land reform that occur in non-violent, yet non-idealized, political practice. All things considered, if EO ‘wins’ for a proposed reform, given prevailing conditions, the door is wide open; if LI wins, the door is almost shut.
If LI and EO are incommensurable but both desirable, what in a particular case determines the strength of arguments for EO or LI? LI and EO each have diminishing returns: the more a society or individual has of one, and the less of the other, the stronger is the moral urgency (and felt need) for the ‘deprived’ desideratum and the less for the ‘ample’ one. Some countries, groups or places feature great respect for LI, but great ascribed inequality. So most people, as non-owners or micro-owners, have great liberty to retain and use individual endowments and assets, but they suffer from severe lack of opportunity inside and outside farming. We suggest that the moral case for LI is weakened, and the moral case for EO-based land reform is strengthened, if current land rights, by historical and international standards, are either/or:
very unequal per person, and leading to extreme inequality of assets, income and opportunity;
inherited, not acquired from savings out of one’s own income from work;
originating in force, fraud, bribery or politicking, especially if recent;
due to colonisation, especially if recent—and above all if colonising groups and their successors, having got rich by land grab and inheritance, later failed to equalise land, non-land opportunity, education or voting rights, as in most of Latin America, but not North America (Torche and Spilerman 2006);
derived from land, rather than improvements or ‘earned’ income from farm enterprise;
shifting income from poor landless workers and small farmers towards rural rentiers who receive such income whether or not they work.
Similarly, land reform violates LI less, to the extent that one or several of the following considerations apply to the case at hand:
land losers or their parents/grandparents acquired that land illegitimately, especially if recently;
land losers receive compensation;
land losers’ alternatives to reform (e.g., risk of land invasion or revolution) are unattractive;
land losers’ alternatives after reform are reasonable;
importantly (and much neglected), horizontal equity is respected: rural landowners bear only a ‘fair’ share of a distributive sacrifice, and are treated no worse than equally rich urban high-income landlords with similar LI; and quite poor rural farmers are not favoured over poorer urban (or rural) labourers.
We do not claim to provide an algorithm that would solve the conflict between the claims of EO and LI in every situation. More modestly, we want to direct attention to the morally relevant considerations that parties involved in deliberation need to take into account in their deliberation. Our goal is to assist deliberators in identifying the reasons that weaken or strengthen for or against EO-based redistributive land reform.
That apart, there are some situations where the politics of potential land reform make legitimacy issues either moot or, if used as criteria for land law or allocation, likely to cause violent conflict among strong claimants to legitimate land ownership or occupancy. In such cases, reducing extreme inequality of land (and hence opportunity) is normally preferable, as a practical criterion for land law, to legitimacy—and redistribution to (claimed) restitution.
In sum, even at the most fundamental level of moral disagreement over land reform, progress is possible if EO and LI can be compared in a case-by-case moral deliberation. We have offered some argument in favour of this approach, which has the advantage of reflecting the non-violent political practice surrounding land reform, rather than being confrontational and requiring that citizens or the government make a mutually exclusive choice between EO and LI.
The ambition of this paper was two-fold. Our first goal was to provide an overview of evidence about facts and consequences of land distribution and redistribution. Our conclusion is that private land redistribution (but emphatically not collectivization) in low-income and most middle-income countries with very unequal land control has not only raised poor people’s share of income, but also accelerated, or at worst not decelerated, growth. It has thus reduced poverty and inequality and increased EO, and food security. Along with increasing farm production capacity, reducing overconsumption, and reducing food waste, land reform offers a major, though too often neglected, leverage for addressing undernutrition and food insecurity in developing countries.
Our second goal was to clarify some crucial issues in farmland access ethics by exploring the roots of the moral case for and against land reform. We have argued that although equality of opportunity and claims of legitimate incumbency based on property rights are incommensurable, the tension between them is not irreducible because these claims can be compared. It is indeed possible to deliberate over the moral permissibility and desirability of land reform on a case-by-case basis. The case-by-case approach is sensitive to the particular features of the context in which the land reform question arises. We have identified some morally relevant considerations that increase or decrease the weight of EO and LI claims. While it is impossible to offer an algorithm that would do away with the uncertainties of moral deliberation, the case-by-case method is not arbitrary. It is responsive to the claims of both camps and offers a way to ponder the reasons in favour and against land reform. The fundamental moral disagreement on land reform is therefore not as intractable as it is usually assumed. This approach has the advantage of being in resonance with non-violent political practice.
Our hope is that this white paper will help us separate genuine moral disagreements from factual disagreement, plain ignorance and the misuse of data. Up to now, the ‘development debate’ has often shown little awareness of philosophical issues, while some debates about the ethics of food and farmland access have neglected facts and relationships among facts, and motivations established by economics, demography and other social sciences. If successful, we have shed some new light on the relationship between farmland access ethics and food ethics in connection to the food security and development debate.
To put things differently, in this paper we have so far identified three types disagreements on land reform:
Factual disagreements: disagreements on facts about land distribution or its effects (e.g., farm size-yield relationships) that often lead to:
Disagreements about consequences of land-related political action;
Ethical disagreements based on the tension between the competing demands of equality of opportunity and legitimate incumbency (property rights).
But to pursue the conversation, we would like to end this paper by suggesting two additional types of disagreements on land reform that we have not yet discussed:
(d) Perspective-dependent disagreements. These disagreements are the consequences of different stances that actors involved in the land debate take.
(e) Ethical disagreements about the potential unfairness of urban-led redistribution of land toward certain populations, in particular the urban poor.
Let us briefly comment on each point in turn. First, perspective-dependent disagreements between stakeholders are difficult to address. This is because neither big landowners nor landless labourers will, even if ‘trying to be fair,’ usually agree either with each other or with academic analysts or ‘objective’ policy advisers. One solution is to put in place a decision procedure in which stakeholders are asked to reason and negotiate ‘under a veil of ignorance’ (to use John Rawls’ terminology) that is, to reason impartially, putting to the side their own identities and interests in order to reach a reasoned consensus. Although procedural solutions are attractive and can certainly play a role in making fair and mutually acceptable decisions about land reform, the Rawlsian form of decision procedure works better in seminars than in reality, and in simulations than in negotiations. Perhaps other types of deliberative and/or participative decision procedures that place (psychologically) more acceptable demands on stakeholders from their own current perspective might be more adequate for making progress on land reform and reaching at least a stable compromise, if not a consensus.
Second, ethical disagreements about the potential unfairness of urban-led redistribution involves real moral disagreement. These disagreements result from the perception that land reform’s benefits are not evenly distributed among the general population, especially the urban poor. Land reforms, like other policies, are in most countries, if not all, made by privileged urban elites. Ought these to redistribute land from middle-income landowners, but sacrifice little or nothing themselves? The contention is then that ethical demands (in particular, those of justice) that are the basis of the case for land reform should consistently be extended to all sources of income, status or power, not just land.
Even here, more facts can reduce the area of disagreement—and permit compromise. Leaving aside the underlying legitimacy, or otherwise, of land redistribution—and also that this may be achieved consensually and/or with compensation—the overriding moral issue about urban-led redistribution of farmland may be seen to be, not the effect on the urban rich, but on the poor. This certainly includes the urban poor (though they remain under 30% of poor people in sub-Saharan Africa and India). Smaller and more equal farms do reduce the proportion of farm output that farmers sell to cities, tending to harm the urban poor (much more than the urban non-poor) by raising its price.32 However, this may be compensated, perhaps more than compensated, as a shift of farmland towards smallholdings increases amounts of farm output produced—and, in respect of urban food provision, shifts composition of output towards food, particularly staples. More generally, if smaller and more equal farms mean higher GDP, the losers from land reform can be better compensated. Finally, more equal small farms raise demand for rural labour, slowing down rural-to-urban labour migration, and so strengthening the market and bargaining position of urban labour.
Agarwal, B. 1994. A Field of her Own. Cambridge: Cambridge University Press.
_______. 2014. “Food Sovereignty, Food Security and Democratic Choice: Critical
Contradictions, Difficult Conciliations,” Journal of Peasant Studies.
Bell, C. 1990. “Reforming Property Rights in Land and Tenancy.” World Bank Research Observer 5(2): 143-66.
Caldwell, L., and K. Shrader-Frechette, eds. 1993. Policy for Land: Law and Ethics, Lanham, MD: Rowman and Littlefield.
Cherniss, J., and H. Hardy. 2013. “Isaiah Berlin.” The Stanford Encyclopedia of Philosophy (Spring 2014 Edition), Edward N. Zalta ed. URL = http://plato.stanford.edu/archives/spr2014/entries/berlin/(accessed August 25, 2014).