There are three hypotheses explaining hunger and malnutrition in the world:

1. there is a global scarcity of food;
2. food is inadequately distributed; or
3. shortages are cyclical, reflecting varying local conditions.

The "strong form" of global scarcity hypothesis (hypothesis 1) is not supported by historical trends or current data. Although the supply of arable land is ultimately fixed and global population is still rising, per-capita food production has continued to increase mainly because of technological improvements (fertilizers, pesticides, genetic improvements, irrigation). Scarcity should imply higher prices, not "shortages." Demand for food is generally inelastic, implying quantity variations should trigger large fluctuations in prices.

Some foods are clearly non-necessities: for example, beef cattle have an 8:1 feed-to-gain ratio, so dietary substitution of grains for beef would permit production of about 8 times as many food calories from the same cropland base.

The big question is whether or not future technological improvements can keep pace with global food demands. There are three areas of possible concern:

1. losses of farmland to other land uses: In the US, the relative value of agricultural land has declined, indicating that farmland loss is not a serious food-security concern (although it may be a valid aesthetic concern): we currently have about 60 million cropland acres idled in the Cropland Reserve Program.
2. rising energy costs: Since energy and capital are production complements, higher energy costs should slow capitalization of agriculture and favor smaller-scale, more labor-intensive farming.
3. environmental damage caused by agriculture: Conventional agricultural practices often generate soil erosion, nitrate and pesticide contamination in groundwater, or other environmental problems; SCS subsidizes soil conservation practices, and environmental externalities are coming under increasing regulation.

The "weak form" of the scarcity hypothesis states that increasing food scarcity causes food prices to increase faster than other prices. Harold Barnett compared ratios of agricultural prices to general wholesale prices for 53 countries, and found only 15 countries in which agricultural prices were rising faster than general wholesale prices.

The maldistribution hypothesis (hypothesis 2) is very plausible: some LDC's with high population growth rates and low per-capita incomes exhibit declining per-capita caloric consumption. The basic policy issue is whether these countries should promote domestic food production (substituting for food imports which use up scarce foreign exchange) or focus on areas of comparative advantage (presumably not food production). Total food self-sufficiency is generally not an economically wise policy objective for these countries.

Food is undervalued in many countries because of market distortions. Government-run marketing boards in LDC's hold food prices down to benefit urban consumers and maintain government popularity in urban areas. But food price controls hurt rural economies, discourage domestic food production, and ultimately increase dependence on food imports. Taxes on cash-crop exports reduce farmer incomes, export volume and foreign exchange receipts. Food stamp programs targeted specifically to poor people are more efficient.

"Green Revolution" technologies designed specifically for LDC's (dwarf wheat, hybrid maize and rice varieties) have increased food productivity dramatically, but have caused some unintended income redistributions as well: wealthier farmers may be better positioned to take advantage of these technologies than small farmers.

Most localized famines in the world today are the direct result of political conflicts, not true shortages. Recent localized famines in Ethiopia, Somalia and some other areas are mainly due to intentional disruption of food supplies by the government in order to "pacify" ethnic minorities or rural insurgencies.

The cyclical hypothesis (hypothesis 3) can be represented via the cobweb model, in which producer responses lag market conditions and cause price de-stabilization. This assumes producers are pretty myopic. Since demand for food is inelastic, low aggregate yields actually increase farm incomes while hurting consumers, while high aggregate yields reduce farm incomes and benefit consumers. Stockpiling can stabilize prices, but doesn't settle the issue of whether food prices should be high (as farmers want) or low (as consumers want) relative to other prices.