Econometric models that predict the global food supply-demand balance have a large influence on policy makers and research investment in the crop and soil sciences. The most influential models predict a scenario of adequate food production capacity and lower real food prices based on continuing gains in crop yield and a substantial expansion of cultivated area. But what if the underpinning assumptions that give rise to this optimistic scenario are wrong? Cereal crops are the foundation of our human food supply, and area cultivated to cereal crops has been declining by 2.1 million ha per year since 1981. Current linear rates of increase in cereal yields are not sufficient to meet projected grain demand on the diminishing area cropped to cereals. Likewise, the econometric models focus on use of grain for human food and livestock feed but do not consider additional demand for production of biofuels and bio-based industrial feedstocks. Given the rapid rise in fossil fuel costs, significant quantities of grain will likely be diverted to produce these non-food and non-feed commodities. Hence, there is a critical need for a large acceleration in the rate of yield gain on existing cultivated land to protect remnant natural ecosystems and marginal soils from expansion of intensive agriculture. In many of the world's major cereal cropping systems, average farm yields will have to reach 90% of the genetic yield potential ceiling. But crop response to applied nutrients typically follows a diminishing return function as yields approach the yield potential limit, which means that nutrient use efficiency decreases and the potential for nutrient losses increase. To avoid wide scale environmental degradation associated with the need to raise yields more rapidly, a well-funded international research effort will be required with an explicit focus on achieving substantial increases in both yield and nutrient use efficiency—especially for N and P. Improving soil quality and the indigenous soil nutrient supply is a key component of the ecological intensification that will be needed. Current levels of funding and the global research portfolio it supports are woefully inadequate to meet this scientific challenge. Global food shortages, high food prices, increased poverty, and severe environmental degradation are possible if we fail to develop technologies for high-yield, high-efficiency cereal production systems appropriate for both large-scale agriculture in developed countries and small-scale agriculture in developing countries.