Food security policies

Bo Döös discusses ways of identifying policies to ensure food security and maintain environmental sustainability.

The author is Director of Global Environmental Management in Stockholm, Sweden.

In view of the rapidly increasing world population, and the declining stock of natural resources, serious concerns have been expressed about the future capacity of the worlds food production system. At present, the developing world contains almost 800 million food-insecure people. Although this number has decreased during the last few decades, and can be expected to continue to decrease in the near future, it will only be at a rate of about one percent per year. Hence, there will continue to be large numbers of hungry or food-insecure people and food demand will exceed supply for a considerable time to come.

The International Food Policy Research Institute (IFPRI) has analysed alternative strategies and policies for meeting food needs of the developing world. A recent report from the Institute examines the effects of changes in policy, technology and life styles. One of the main conclusions is that even rather small changes in agricultural and development policies and investments can have wide-reaching effects on the number of poor and undernourished people around the world.

This article, and the study it is based on, is concerned with the means for selecting an optimal combination of policy actions in developing countries that satisfy the objectives of food security while maintaining environmental sustainability.

The inevitable uncertainties in the projections of food demand and availability over coming decades mean that we are not in a position to state categorically that there will be widespread shortage of food and an unavoidable need for costly policy actions.

For this reason, I consider not just the selection of policy actions that are directly aimed at satisfying the stated objectives. I also review policy actions that are directed at other global issues, but which may also impact food security positively. This may permit the need for the identified policy actions to be presented more convincingly.

Policy needs

In attempting to develop and implement policy actions aimed at improving agriculture, we must confront the following questions. What future capacity of the global food production system will be sufficient to meet demand? What future levels of food security will policy changes be required to ensure?

Although considerable efforts are being devoted to developing models for prediction of the global demand for and supply of food, our abilities to make reliable predictions for the next few decades are very limited. The main difficulty is modelling the driving forces such as socio-economic and political factors and realistically accounting for the increasing human-induced environmental stresses, such as increasing frequencies of floods and drought, heat waves, and so on.

There are also other limits to predictability. For example, ambitions to design a truly comprehensive food supply and demand model may lead to the inclusion of processes for which understanding is incomplete and, hence, serious errors.

In light of these uncertainties, we expect governments will be reluctant to commit substantial financial resources for financing policy actions that they believe may not be needed. Some predictions of future food production appear to be overly optimistic. The reason for such optimism is often that the predicted changes are evaluated on the basis of what is theoretically possible, rather than what is likely to happen.

There will always be those who deny, or deliberately underestimate the risks, or argue that the scientific base is too weak. These people may argue in the name of scientific prudence even when their real reasons are short-term economic concerns. This problem was recently illustrated in the ongoing attempts to reach international agreement on the reduction of the emission of greenhouse gases and thereby to reduce global climatic change.

Policy objectives

In developing a strategy for implementing policy actions aimed at meeting future food needs, there are basically three objectives.

  • Sufficient increase of food production the supply of food should meet the likely increase in food demand.
  • Food security for all access to food must be improved and the number of undernourished and malnourished children and adults reduced in the developing world.
  • Environmental sustainability the unnecessary loss of the limited world resources of soil, biota and water must be avoided.

The objectives of policy actions are, therefore, not limited to merely augmenting agricultural production. They also include access to the supply of food and the availability of a more healthy diet. Moreover, the rate of use of the natural resource bases should not exceed the rate at which nature can produce them.

A large set of policies exist from which any individual country can choose in achieving the three basic objectives. However, it must be emphasized that the need for particular policy actions is very different in different countries.

Figure 1: Examples of policies that can contribute directly to achievement of the three objectives listed in the text, either directly (thick arrows) or indirectly (thin arrows).

In Figure 1, some of the more important policies have been identified. These examples encompass the five sectors of investment drivers irrigation, rural roads, education, drinking water supplies, and agricultural research identified by IFPRI as being the most important drivers in their International Model for Policy Analysis of Agricultural Commodities and Trade.

In addition, three categories of policy actions directed at improving environmental sustainability have been included in the figure. The figure also identifies policies that can be justified even if there is no concern about the sufficiency of future global food production.

Increasing food supply

The first objective, as shown in the figure, is to reduce the gap between demand and the supply of food. The following are three examples of policy actions.

Irrigation. This sector ranks highest among the projected investment requirements in a baseline scenario presented in an International Food Policy Research Institute study. This study does show, however, that low rates of return have been experienced in many recent irrigation projects. It is considered that the opportunities for expansion are mainly concentrated in south Asia, India and Latin America.

Fertilizers. Considerable opportunities to increase food production still exist in many developing countries by applying more fertilizers. However, it must be taken into account that the gain in applying more fertilizers is much less than it was in the past. It should also be recognized that there can be demands for a more restrictive use of fertilizers in view of their potential negative effect on the environment.

Research Products. Recent experience indicates clearly that strong returns from investments result from agricultural research and management improvements, such as developing locally-appropriate crop varieties or providing farmers with better extension services.

Improving food security

The number of people who live with food insecurity has been decreasing during the last few decades, despite the rapid growth of the worlds population. Nevertheless, with about 800 million people still without enough to eat, the decline must be considered insufficient. It is abundantly clear that sustainable food security for all is impossible by 2020 if business is conducted as usual.

According to Food and Agricultural Organization projections made in 2000, the goal of reducing the number of food-insecure people to 400 million by 2015 will not be reached until 2030. IFPRI projections suggest similarly slow progress in reducing child malnutrition. By 2020, 132 million children under the age of six years, that is, one out of every four children, may be malnourished. There is, therefore, an urgent need to identify, prioritize and implement policy actions that can contribute to improving food security and reducing malnutrition.

The following are six examples of areas where policy actions would be most effective.

Health care. It has already been emphasized that it is important that policies are directed at improving health and nutrition. Good health and nutrition are the means as well as the ends to eradicating poverty and achieving broad-based development.

Poverty. Poverty alleviation is essential, for poverty ruins lives and undermines development, and environmental and political stability.

Infrastructure. A review of investments in 14 developing countries has revealed wide disparities in infrastructure availability between rural and urban areas. An increase of rural roads helps alleviate poverty and thereby increases food security. The need for this type of policy action is particularly important in Sub-Saharan Africa and Latin America. Policy action on infrastructure also contributes to increasing crop yields and opportunities for cropland expansion.

The role of women. Policy makers must recognize that women are major contributors to agriculture and play a prime role in ensuring food security. In a 2001 report, the United Nations Population Fund presented seven policy areas involving increased understanding of womens importance, all of which required immediate attention.

Education and clean water. Policies aimed at improving education and providing access to clean household water are particularly important in reducing the number of diseased and malnourished children.

Agricultural trade. As pointed out in 2002 by the International Policy Council on Agriculture, Food and Trade, the developed countries need to do more to open agricultural markets to developing countries in order to increase their agricultural production and thereby reduce poverty and hunger.

It must be emphasized that implementing these examples of policies would not only contribute to improving food security, but they would be totally justified in their own right.

Increasing environmental sustainability

There is a considerable loss of food production caused by different types of environmental degradation. Three particular problem areas that need policy actions implemented are the issue of soil degradation, loss of cropland and the issue of climatic change.

There is a growing concern that ongoing soil degradation represents a serious threat to food production in the developing world. In a paper presented by IFPRI in 1999, several problems of soil degradation for food security were considered. In order of global policy priority, they include:

  • improving soil quality in densely populated marginal lands;
  • improving system- and farm-level water management regimes and investing in proper drainage systems;
  • reducing degradation of high quality rainfed land by improved integration of suitable technologies;
  • preserving agricultural land quality in urban and pre-urban agricultural lands; and,
  • limiting environmental degradation in marginally-productive lands caused by intensive agriculture.

Each year a considerable amount of cropland is lost by conversion from rural to urban use in order to serve the rapidly growing population. This considerable loss caused the United Nations Conference on Environment and Development to launch in 1991 a comprehensive programme aimed at developing an Integrated Approach for the Planning and Management of Land Resources. Another point which should be noted is that the loss of cropland will likely be replaced by land reserves with poorer productivity.

Recent regional, human-induced climatic changes, particularly temperature increases, have already affected many physical and biological systems. In the Third Assessment Report for the Intergovernmental Panel on Climate Change, it was observed that the projected adverse effects would include:

  • a general reduction in potential crop yields in most tropical and subtropical regions for most projected increases in temperature;
  • a general reduction, with some variation, in potential crop yields in most regions in mid-latitudes for increases in annual-average temperature of more than two degrees Celsius;
  • decreased water availability for populations in many water scarce regions, particularly in the subtropics; and,
  • widespread increases in the risk of flooding for many human settlements as the lessened rainfall is increasingly delivered in a few large storms.

A large number of possible policies, measures and instruments do exist for countries to limit their emissions of greenhouse gases. However, the emissions reductions agreed on by governments so far are about one order of magnitude below what is required to have the desired effect, stabilization of the atmospheric concentration of greenhouse gases below present day levels.

Policy implementation

Any selection of policy actions must include consideration of the generally long lead-time before they become effective. Time lags are caused by three principal factors.

First, there is the need to develop a firm scientific basis for certain policies. To achieve a consensus within the international scientific community is very time consuming. Global change science provides a recent example. Fossil fuel burning effects on climate were first suggested by Svante Arrhenius in 1896 and calculated by Guy Callendar in 1938, with large-scale research beginning in the 1970s. Consensus still was not reached until early in the 21st century.

Second, the complex and tedious process of negotiating national and international agreements on policy actions can create a time lag. Consider the Kyoto Protocol to the climate treaty. Negotiations were completed in 1997 but the agreement is still not ratified at the time of writing in 2003. As already observed, governments often hesitate to accept the scientific rationale for implementation of costly policy actions. Their reticence may also produce underestimates of the response actions required.

Finally, there is the time required to accomplish widespread implementation of the approved policy actions due to limitations of financial resources, human behaviour patterns and other sociological processes.

Two approaches

It is now time to consider two different approaches in selecting policies aimed at the given objectives. Each demands a timely implementation if it is to minimize lags. The basic difference between these approaches is the differing probabilities of financing their implementation. In order to illustrate the basic philosophy behind these two approaches, the available policies that contribute to the stated objectives have been divided into two different categories.

Portfolio A includes policy options that have a direct positive impact on one or more of the three objectives. It, therefore, includes policies that are specifically designed to augment food production, such as by expansion of irrigated land or making use of new crop varieties, or to increase reliability of food distribution, such as increased transportation infrastructure. However, as pointed out, governments probably will be reluctant to implement such policies, if viewed as costly, due to uncertainties in the predictions of global food production.

Portfolio B includes policies that are related to other global issues but can also contribute to achieving food security and environmental sustainability. Examples include policies aimed at reducing poverty and improving education in the developing world. This portfolio includes policies aimed at a substantial reduction of the emission of greenhouse gases, consistent with the objective of the United Nations Framework Convention on Climate Change.

A straightforward approach is to choose policies that serve the three objectives stated above in the most efficient way and to assume that their implementation can be financed. Thick lines in the figure on page 18 connect such policies to the objectives listed there. Many of these policies also contribute to other objectives such as increasing the role of women. These achievements in turn contribute, for example, to increasing food production as indicated by the thin lines in the figure.

An alternative approach is to choose policy options that a) are aimed at other global issues, but at the same time also contribute to the stated objectives, and b) are considered politically feasible and have a high probability of being accepted and financed. In support of such an approach, four primary views should be noted:

  • The first is that in addition to policies that are directly aimed at improving yield and expanding crop area, agricultural and rural growth can be facilitated in numerous other ways (for example, through policies that are aimed at improving rural infrastructure, education or health care).
  • The second is that agricultural growth is a major contributor to economic growth in many developing countries and so assists in reducing poverty.
  • The third view is that empirical evidence shows that the most efficient way to reduce under-nutrition and malnutrition is through economic growth focused on the poorest.
  • The fourth view is that one of the major causes of environmental stress in the developing world is poverty and one of the major causes of poverty is environmental stress.

An optimum policy combination

Now I consider assembling a combination of policies that satisfy the stated objectives in an optimum way. To accomplish this there are certain assumptions which are required to be made particularly regarding basic uncertainties. For each individual country or region, assumptions need to be made with regard to the likelihood of attaining the objectives without policy action being taken. Present projections exhibit very different results about the extent to which food supply will meet needs. The magnitude of the impacts of the greenhouse problem will also need to be estimated.

Assumptions also need to be made in regard to the effectiveness of policy actions. In general, the present predictability of policy impacts is very limited. Consequently, it is difficult to judge how effective the various policies will be in achieving the objectives. This implies that, at present, we understand too little to design an ideal combination of policies.

A final important assumption we must make is in regard to the probability of funding the required policies. The choice of a set of policies to achieve established objectives depends, to a large extent, on the probability of funding the individual policies. For example, policies related to education and the role of women might be given considerably higher priority than policies aimed at environmental sustainability.

When attempting to make an optimum selection of any relevant policies, several different approaches could be used. I propose a relatively simple, iterative method for selecting a nearly optimum combination of policies that satisfies the given objectives. The procedure for the selection of the desired combination of policies consists of the following steps.

  • First, identify all policies within both portfolio A and portfolio B that can contribute significantly to the achievement of the objectives, either directly or indirectly.
  • Second, determine the following quantities for each of the major regions of the developing world: food demand; food supply without policy actions; and food supply obtained for each one of the identified policy actions.
  • Third, make as realistic estimates as possible for each individual policy regarding: the extent to which it contributes to achieving the stated objectives; the investments required for implementing the policy for a given impact and the probability that the policy can be funded and implemented.

Table 1: Examples of policies that contribute directly (Portfolio A) or indirectly (Portfolio B) to the three basic objectives. + indicates a positive impact and - indicates a negative impact by increasing policy action.

Policies relating to:

1: Increase food production

2: Improve food security

3: Reduce environmental degradation

Portfolio A


+ major


- medium

New crops

+ major



+ major


Clean water


+ major


Health care


+ medium


Rural finance


+ minor


Soil degradation

+ major

+ minor

+ major

Loss of cropland

+ major

+ minor

+ medium

Greenhouse gas emissions

+ medium


+ minor

Portfolio B


+ major

+ major

+ major

Role of women

+ major

+ major

+ major


+ minor

+ major

+ minor

Poverty reduction

+ medium

+ major

+ minor

  • Fourth, based on the outcome of the previous step, choose a preliminary subset of policies that can provide a significant contribution to achieving the objectives (see Table 1).
  • Fifth, based on the selected policy options, project the specific values that represent the three objectives. It is recognized that these projections will be somewhat inaccurate.
  • Sixth, assess to what extent the selected combination of policy actions satisfy the three objectives. If the outcome of this assessment is not satisfactory, a new set of policy options must be chosen, and new projections undertaken. This implies a repetition of steps 4, 5 and 6.
  • Repeat until an acceptable combination of policy actions has been obtained.

This approach has been applied to identify the types of policies required to ensure that, during the next few decades, the three important objectives are met:

  • that food supply meets demand;
  • that food security is available for all; and,
  • that no unnecessary losses of natural resources take place.

A way forward

One plausible outcome of this approach suggests that both Portfolio A (direct) and Portfolio B (indirect) policies need to be deployed to meet the three objectives and that a substantial reduction in the number of food-insecure people requires extensive use of Portfolio B policies, such as improving health care and education.

Although the study cannot be considered comprehensive, certain conclusions can be drawn.

It is clear that there exists a wide spectrum of policies that contribute either directly or indirectly to the objectives. Also, some of these policies contribute to more than one of the stated objectives.

The selection of an ideal combination of policies is possible, but is handicapped by our limited ability to reliably predict future food production and the state of the environment. These uncertainties make it difficult to provide convincing arguments concerning the need for policy actions and investments aimed at increasing food production.

The choice of an ideal policy combination is also hampered by current limitations in knowledge about the impacts of the various policy actions. Hence, an approach to improve food security and environmental sustainability indirectly through policy actions directed at other global issues is likely the best. Such no regrets policies can be more easily accepted and financed by governments. It must be remembered, though, that policies aimed at serving one objective can negatively impact another.

Finally, early identification of the required policies is essential to allow for the long periods often required for policies to be clearly defined, agreed upon, financed and implemented.

Further information
Bo Döös, Packhusgrand 6, SE-111 30 Stockholm, Sweden. Email: Web:

Full report
A full report on this work by Bo Döös and Allen Solomon is available in Adobe Acrobat format (file size cc. 600kb).