1.74 | Fall 2020 | Graduate

Land, Water, Food, and Climate

SECTION 4 | Reconciling Demand and Supply: Options

Overview

What are the primary options for reconciling food demand and supply? What are their advantages and disadvantages?

Our discussion so far provides background we need to evaluate options for reconciling food demand and supply. The available options are constrained by physical factors such as the land and water available for growing crops and by human factors such as income, dietary preferences, and attitudes about pesticides or genetically engineered crops. Before we consider particular proposals for achieving sustainable food security it is helpful to step back and review some of issues revealed in our readings.

We have seen that increasing production beyond present levels could conflict with the need to maintain a sustainable agricultural system that functions well over the long term. In particular, the inputs and practices needed to increase yield have already had, and will likely continue to have, adverse environmental impacts that threaten the land, water, and soil resources we need to grow food. Of course, we would like to have a win-win solution (sustainable intensification?) that is both sustainable and sufficiently productive to support our growing population. But that may not be so easy to achieve.

We have seen that food security varies greatly with location, time, and income so that, while global analyses give some overall perspective, they are not sufficient to assess the feasibility and desirability of alternative food security proposals. For example, we have seen that current global food production appears to be enough to feed everyone on the planet a nutritious diet. But several hundred million people are still chronically malnourished.  There are many more examples where geographical and economic variability are critical to an assessment of food security problems and solutions.

It is worth highlighting some critical regional differences that affect food security. Land suitability (e.g. soil, terrain, and growing season length) and water availability vary greatly, with the best conditions generally found in areas that have been cultivated for centuries, primarily in the northern hemisphere. Prospects for new cropland opening up are limited, although climate change could shift the areas that are most suitable. In general, tropical and subtropical areas in the global south appear more likely to be adversely affected by climate change but some northern hemisphere areas such as the Mediterranean and western North America could also suffer. The level of economic development in different regions also has an important impact on crop production and food security. Africa stands out for its low per capita income, higher malnutrition rates, and lower life expectancies as well as its low crop yields. But there are pockets in all regions where the food supply is currently insecure or is threatened by poverty, political instability, and/or climate change.

Regional diversity has led to differences in the way that farming is conducted. It is helpful to distinguish three broad crop production systems that have global extent 1) commercial export-oriented, 2) small high-input, and 3) small low-input farms (see S13 for a quick comparison):

Commercial export-oriented farms tend to be in the areas most favored for crop production, especially in the more economically developed countries of North America, southern Latin America, Europe, and Oceania. Farms in major food exporting countries are generally large, employ small fractions of their national populations, and provide farmer incomes that compare well to incomes in other occupations. These farms supply most of the essential food needs of their regions while also exporting enough food (calories) to feed as much as one third of the global population (see Class 10). They tend to use high levels of inputs (e.g. water, nutrients, pesticide), rely on modern high-yield cultivars, have better market access, and are more mechanized and less labor intensive than farms in the other two groups.

Small high-input farms, especially small farms in Asia, have benefitted from the twentieth century Green Revolution, which introduced new cultivars selectively bred for robustness and high yield. Farms that rely on these cultivars typically use high levels of fertilizer and pesticides and achieve yields nearly as high as commercial farms. However, they tend to be much smaller, more labor intensive, and provide less income for their operators than farms in the first group. The reason for the persistence of small farms in Asia is debated. Some researchers suggest that there has not yet been sufficient time for a transition to large farms such as those that occurred in Europe and North America, when farm labor productivity increased and many rural workers left for urban jobs. Others feel that small high-input farms survive in Asia because they work well and are appropriate for local conditions (see Class 7).

Small low-input farms generally do not follow the high input/high yield model of the other two groups. Farms in this third group often do not have ready access to essential inputs, are labor intensive, have low yields and revenues, and are limited by inadequate infrastructure and market access. Poor smallholder farms prevail in sub-Saharan Africa but are also found in parts of Asia and Latin America, as discussed in Class 7. Their future is uncertain and there are many who believe that they cannot survive.

Each of these farming systems currently feeds a sizable fraction of the global community. They tend to cluster in particular regions, as revealed in part by the striking field size map shown in Figure S10. All three systems are relevant to our consideration of future food production options in a world with a few billion more people, higher per capita consumption, and a changing climate. Each provides a possible model for future agriculture and, moreover, a place to start when considering other models that are different from any of these present-day options.

In Section 4 we start by reviewing, in Class 11, the twentieth century Green Revolution that was instrumental in creating the small high-input Asian model of agriculture. Then we consider whether a repeat or extension of the Green Revolution could provide a sustainable global food security solution for the 21st century. In Class 12, we examine the primary ideological alternative to the Green Revolution model—agroecology. This option advocates reliance on ecological principles and a suite of smart low-input (or alternative input) farming practices. In some ways, agroecology can be viewed as an improved version of the small low-input model production system. As we explore the common areas and differences between the Green Revolution model and the agroecological model we will also be looking for hybrids and totally new options that may suggest a way to achieve both food security and sustainability.

It should be easy for you to identify the advocates for different perspectives in our readings. There are relatively few papers that give equal weight to food security and sustainability. The debate over the merits of the Green Revolution vs. agroecology is linked to debates we have already encountered over genetically engineered crops, sustainable intensification, and the desirability of encouraging smallholder agriculture. Although we have read only a few selections from the large literature on these topics our examples are sufficient to illustrate the high degree of controversy found in discussions of food security. This applies to the scientific community as well as the economic and policy communities. We will need to recognize this controversy when discussing a path forward in Section 5.

Section 4 Class Topics

Class 11: Green Revolutions Past and Future
Class 12: Agroecology

Section 4 Supporting information (SI)

S13. Comparison of Three Crop Production Systems

In Class 11 we examine more closely the high input agricultural systems that now prevail in somewhat different forms in developed countries and a large part of the developing world. These systems, both the large and small farm versions, rely on modern high yield cultivars, extensive irrigation, and synthetic fertilizers. Our discussion builds on readings covered earlier, especially in Section 3, as well as new readings that focus on the feasibility of further increasing production with the high input model.

© Springer Nature. All rights reserved. This content is excluded from our Creative Commons license. For more information, see https://ocw.mit.edu/help/faq-fair-use.

In the mid-twentieth century the high input modern cultivar approach widely used in European and North American agriculture was extended, with considerable success, to Asia and parts of Latin America. This so-called “Green Revolution” was driven by selective breeding programs that developed hardier, higher yielding varieties of wheat and rice (see attached sketch from Khush (2001)). As the new varieties were adopted they increased cereal production and provided substantial improvements in food security. The short article by Hazell (2002) summarizes the history and accomplishments of this transformation.

The Green Revolution introduced environmental problems, as well as greater food security, to the countries where its methods were adopted. Many of these problems were related to the increased use of fertilizers, irrigation, and pesticides that were needed to fully realize the higher yield potentials of the new varieties. This increase in inputs, which was often not carefully controlled, led to the adverse environmental impacts now widely associated with modern agriculture and discussed in Class 6. The paper by Pimental and Pimental (1990) provides a brief summary. More details on relevant economic issues, environmental impacts, and farm consolidation are provided in a longer optional reading by Hazell (2010). The optional reading by Khush (2001) provides background on the genetic methods used to breed Green Revolution cultivars.

Green Revolution technology transformed much of smallholder agriculture in India and China from the small farm/low input model to the more productive small farm/high input model introduced in the Section 4 Overview. The transformation was most successful in areas with sufficient water to support irrigation systems and with ready access to fertilizers and locally appropriate seeds. Although efforts were made to introduce the Green Revolution to Africa, these were less successful, leaving a sizable and rapidly growing population with the lower yields, less reliable production, and poverty generally associated with low input smallholder farms. This aspect of the Green Revolution is briefly discussed in Carr (2001) in Class 7.

Does the success of the twentieth century Green Revolution imply that the best way to achieve global food security is to extend intensification more widely, perhaps by increasing production in the major exporting nations, but certainly by increasing it in Africa and the parts of Asian and Latin America that have been left behind? That will require resolution of the issues raised by Carr, including development of irrigation infrastructure and more ready access to inputs and markets.

The possibility of further intensifying the Green Revolution approach is addressed in the widely cited paper by Cassman (1999). He considers the possibility of raising cereal production by increasing potential (unstressed) yield (Class 5), controlling and improving soil quality (Class 6), and using precision agriculture to improve the efficiency of nitrogen application (Class 9). In each case, his conclusion is that small rather than revolutionary improvements are most likely. Cassman mentions that new genetic engineering methods could conceivably have a major impact if they can increase yield potential significantly beyond what has already been achieved with traditional selective breeding. However, the primary contribution of these methods so far has been to bring actual yields closer to existing potential yields by reducing stress, primarily from losses to pests.

Soil quality improvements could also be beneficial but require more than just additional nutrient application, as illustrated in the paper’s examples of recent yield declines in irrigated rice systems. Post-Green Revolution yield declines have also been observed in other settings (see Class 5). Although precision agriculture holds promise for improving the efficiency of nutrient and irrigation water application it focuses on bringing actual yields up to potential levels by reducing stress, rather than on increasing potential yield. The overall implication of Cassman’s analysis is that site-specific measures to close yield gaps could make substantive differences where actual yields are much lower than yield potential (e.g. in areas where low input smallholder agriculture still dominates). But there is relatively little reason to believe that there will be another round of dramatic increases in cereal yield potential by mid-century. This raises serious questions about the feasibility of feeding an additional 2 or 3 billion people with intensified Green Revolution methods.

In food security, as in climate change, it seems that the most realistic options are those that combine a number of measures that may have relatively small effect in themselves but may have significant impact when taken together. We return to this idea in Section 5, after a look at the Agroecology alternative to the Green Revolution.

Required Readings

Background on the Green Revolution

  • Peter B. R. Hazell. 2002. “Green Revolution: Curse or Blessing? (PDF)” No. REP-9450. Washington, DC (USA), IFPRI. 3 p.
  • D. Pimentel and M. Pimentel 1990. “Comment: Adverse Environmental Consequences of the Green Revolution.” Population and Development Review, 16, 329–332.

Possibilities of Further Extending the Green Revolution’s Intensification Approach

Optional Reading

Green Revolution impacts

Genetic Background for the Green Revolution

Discussion Points

  • Based on what you have read in class, is it likely that large-scale high input agriculture will spread throughout the areas where it is feasible and feed the rest of the world through exports? In this model, which is believed by some to be inevitable, small farms, either high or low input, will essentially disappear, replaced by a more industrialized global food system.
  • Do you agree with Hazell and others who believe that the Green Revolution’s environmental problems were due to farmer limitations and policy failures rather than any intrinsic deficiency in Green Revolution methods?
  • Although the Green Revolution may have its limitations, it seems to have been successful at increasing national production. Should we be looking for an alternative when considering the future or learn from the past and improve the basic model so production can be increased further and undesirable side effects minimized?

In Class 12 we consider agroecology, which is as much a philosophy and a movement as a set of techniques for growing crops. As the name implies, this perspective views a farm or larger agricultural area as an ecosystem to be managed for human benefit, primarily food production. The intent of the agroecological approach is to rely on practices that work in harmony with natural processes. This may or may not increase yield over conventional (aka Green Revolution) approaches and may or may not alleviate particular environmental problems. “Agroecological” is a broader term than “organic” since organic practices are more regulated and often more restrictive. However, we generally use the terms interchangeably in this discussion.

The practices most commonly associated with agroecology are summarized in the reading by Wezel et al. (2014), which is oriented towards conditions in temperate developed countries. The paper by Pretty (2008) takes a broader global view and provides more discussion of yield and environmental aspects. Both papers are longer than necessary but can be read efficiently. Wezel et al. conveniently summarize agroecological practices in Table 3, which gives the essence of the paper. These practices are divided into “Efficiency increase and substitution” and “Redesign” categories and are discussed in more detail in Sections 3 and 4 of the paper. Wezel et al. note the increased labor requirements of some agroecological practices, such as intercropping and non-chemical pest management. This is an important issue, especially in areas with labor shortages or high labor costs.

Pretty provides in his Section 5 a short list of practices similar to those identified by Wezel et al. and then discusses some of the possible impacts of these practices in Sections 6-8 of the paper. He ends with the somewhat surprising quote, considering the paper’s optimistic assessment (in Section 6) of agroecological yield performance in developing countries:

“ … In this context, it is unclear whether progress towards more sustainable agricultural systems will result in enough food to meet current needs in developing countries, let alone future needs after continued population growth (and changed consumption patterns) …”

This quote recalls the analysis of van Ittersum et al. (2016) from Class 7, which suggests that some sub-Saharan African countries may not be able to meet their food needs even if their yield gaps are greatly reduced. The optional FAO (2013) report on agroecology is a collection of many papers on specific topics that are beyond the scope of our discussion.

Agroecology (especially organic production) is generally associated with reduced inputs of pesticides and synthesized inorganic nitrogen fertilizers. However, an agroecological farm may still consume inorganic nitrogen (e.g. in manure) in quantities that are comparable to the nitrogen used by a conventional farm that grows the same crops. A similar comment applies to irrigation water. Further details on the external input requirements and environmental impacts of agroecological farming are provided in the optional papers by Clark and Tilman (2017), Cassman at al (2002), and Robertson and Vitousek (2009).

Clark and Tilman summarize results from life cycle assessments of a limited number of environmental indicators in a direct comparison of conventional and agroecological production systems. They also compare these indicators for different food groups. They conclude that the two production systems have similar nutrient-related impacts on the environment. They also conclude that meat consumption has significantly more impact on their environmental indicators than the production system. Cassman et al. and Robertson and Vitousek present careful analyses of agricultural nitrogen inputs and related environmental impacts. These relatively long papers are worth reading for the useful information they provide on the origin and fate of nitrogen in its many different forms. They confirm that differences in the way nutrients are processed in the two production systems are subtle and still not well understood. In particular, it is not clear that natural alternatives to synthetic nitrogen fertilizers could meet current, let alone projected, global production requirements.

© FAO. All rights reserved. This content is excluded from our Creative Commons license. For more information, see https://ocw.mit.edu/help/faq-fair-use.

These readings leave us with the question of whether the agroecological approach can achieve the crop yields needed to meet the food needs of up to 10 billion people later in this century. This scale-up question is rarely addressed in publications that advocate agroecological methods. There is no doubt that individual farms can be operated with minimal synthetic nutrient and pesticide inputs if enough effort is expended and increased yield is not a high priority. An example is described in the attached chart that summarizes a 21-year comparison of agroecological and conventional methods at the DOK experimental farm in Switzerland (UNFAO, 2014; Fliessbach,  2007). However, it is still unclear whether agroecology can be a feasible alternative to modern Green Revolution agriculture, which currently produces a major part of the global food supply.

Seufert and Ramankutty (2017) provide a reasonably objective meta-analysis of agroecological performance based on studies carried out mostly in developed countries. The results are summarized in multi-dimensional diagrams that compare agroecological and conventional production systems. The paper states that agroecological (or organic) yields are generally 20% lower than those from conventional systems, with significant variability depending on the crop and on specific management practices. The rankings for other performance measures vary. Crop-specific details of the analysis are provided in the paper’s Supplementary Information and more information on organic yields is given in the optional reading by Seufert and Ramankutty (2012).

An important qualification of the Seufert and Ramankutty work is that much of the supporting data are from developed countries where conventional yields are significantly higher than in most developing countries. Also, reliance on data from certified “organic” farms, as compared to the broader set of agroecological farms, may skew the results to lower yields (because, for example, synthetic fertilizers and chemical pesticides are not allowed in organic production). In any case, agroecological yields that are similar, or even 20% lower, than conventional yields in developed countries could still provide a substantial increase over current production levels in developing countries.

Although the papers discussed in this class do not give a definitive assessment of agroecosystem performance they do suggest that flexible agroecological methods may be beneficial in developing countries in Asia, Africa, and Latin America, where input cost, preservation of soil quality, and resilience are especially important and labor cost is less of an issue. Some agroecological practices may also be attractive for larger farms in exporting countries if these practices can provide substantive environmental benefits without major reductions in yield. A tangible example is the adoption of no-till agriculture in the drier portions of the US Great Plains. This is motivated by a distinct improvement in water retention when tillage is reduced.

It may be that the most important benefits of agroecology are its emphasis on efficient resource use, biodiversity (see photo below), non-chemical pest management, and improvement of long-term soil quality. All of these contribute to the sustainability and resilience of the overall agricultural system.

© Ravi Prabhu. All rights reserved. This content is excluded from our Creative Commons license. 
For more information, see https://ocw.mit.edu/help/faq-fair-use.

Required Readings

Summary of agroecological practices

Comparison of Agroecological and Conventional Performance

Optional Reading

Agroecology Overview

Nitrogen in Agriculture

Comparison of Agroecological and Conventional Crop Yields

Discussion Points

  • Do you think that agroecological and/or organic methods could feed the current or projected 2050 global populations with the current FAOSTAT global average diet? How might your answer change with a different diet?
  • What was your impression of organic agriculture coming into this class? How has it changed, if at all?
  • Elaborate on the pros and cons of donors and governments encouraging agroecological methods in developing countries, especially in Africa. Should we apply different criteria in these countries (as opposed to developed countries) when assessing the desirability of agroecology? Is agroecology a luxury for the rich or an option that could be very successful in the developing world because it builds on tradition and requires less capital?

The three crop production systems described in the Section 4 Overview each include farms found in many different geographic regions. This makes it difficult to compile statistics that characterize each system as a whole. However, it is possible to identify a representative region or country in each production system so that the three systems can be conveniently compared.  This is feasible because the differences across the systems are so significant.

Table S13.1 below lists a number of different demographic, agricultural, and socioeconomic indicators for the three production systems, based on the following representative countries or regions:

Commercial export-oriented farms: USA
Small high-input farms: China
Small low-input farms: sub-Saharan Africa

The statistics in the table apply ca. 2010 and are taken primarily from FAOSTAT, periodic USDA statistical reports (USDA ERS and NASS), IFAD (2010), Wu et al. (2018), and Lowder (2015).

Table S13.1

                                                        USA                China             Sub-Saharan Africa

Population
Total population (million):                     300                  1350                777

UN projected 2050                               380                  1410                2100
   Population (million)
Agricultural population (million):           5.1 (1.7%)       842 (62%)       432 (56%)

Fertility (births per woman)                  1.9                   1.7                   5.2

Land
FAO arable land (million ha)*               156                  106                  165

Fraction cropland on                            —                    80%                 66%
    farms < 5ha
Fraction cropland on                            91%                 –                      –
    farms > 100 ha                                                                                        

Inputs
Cropland irrigated                                14%                 36%                 2%

N fertilizer use (106 tonnes                  12 (0.1)           31 (0.3)            0.5 (0.003)
   and tonnes arable ha-1)
N fertilizer production (106 tonnes)        9                    36                    1

Cereal production
Cereal production                                434                  554                  131
Cereal exports (106 tonnes)                64 (15%)         21 (4%)           2.7

Cereal imports  (106 tonnes)               10 (2%)           1   (0.2%)        30 (23%)

Socioeconomic
GDP per capita                                    $ 51,000          $ 1894             $ 632

Typical annual farm                              $ 60,000          —                    —

   household incomes                           $200,000

Rural residents                                    —                    35%                 87%
   living on less than $2 per day:
Infant mortality                                     6                      19                    89
   (per 1000 births)                                                                              
Life expectancy (years)                       79                    75                   51

Fraction undernourished                     3%                   9%                  30%

* Temporary food and fodder crops

A review of these statistics suggest:

  • The US and China have much lower birth rates than sub-Saharan Africa.
  • US farms are much larger than in the other two regions and are operated by a much smaller fraction of the total population.
  • The US and China have much more irrigated land and produce and consume much more fertilizer than sub-Saharan Africa.
  • The US exports a significantly larger fraction of its cereal production than the other two regions.
  • Sub-Saharan Africa is much more dependent on cereal imports than the other two regions.
  • Sub-Saharan Africa farmers are much poorer and more subject to health problems than the other two regions.
  • China has been able to achieve much better income and health results than sub-Saharan Africa, even though a larger fraction of its cropland is on small farms. Does this suggest that it could be possible for sub-Saharan agriculture to prosper with primarily small farms? Or should it emulate the US large farm model rather than China’s?

References

IFAD International Fund for Agricultural Development. 2010. “Rural Poverty Report 2011 - New Realities, New Challenges: New Opportunities for Tomorrow’s Generation.” Annexes 1 and 2. IFAD, Rome

Sarah K. Lowder, Jakob Skoet, Terri Raney. 2016. “The Number, Size, and Distribution of Farms, Smallholder Farms, and Family Farms Worldwide.” World Development, 87, 16–29.

USDA ERS and NASS. 2012. “Agricultural Resource Management Survey.” (Accessed 22 July, 2020).

Yiyun Wu, Xican Xi, et al. 2018. “Policy Distortions, Farm Size, and the Overuse of Agricultural Chemicals in China.” Proceedings of the National Academy of Sciences, 115, no. 27: 7010–7015.

Course Info

As Taught In
Fall 2020
Level
Learning Resource Types
Lecture Notes
Instructor Insights