AbstractWhile annual precipitation in much of the US Corn Belt is likely to remain constant, atmospheric vapor pressure deficit (VPD), the driver of crop water loss (evapotranspiration; ET), is projected to increase from ~2.2 kPa today to ~2.7 kPa by mid‐century primarily due to the temperature increase. Without irrigation, it has been hypothesized that the increase in VPD will create a ceiling to future increases in maize yields. We calculated current and future growing season ET based on biomass, water use efficiency, and the amount of yield these levels of ET would support for maize production in the Midwest USA. We assumed that the production of more grain will necessitate a proportional increase in the production of biomass, with a corresponding increase in ET. Here we show that as VPD increases, maintaining current maize yields (2013–2016) will require a large expansion of irrigation, greater than threefold, in areas currently supported by rain. The average predicted yield for the region of 244 ± 4 bushels/acre (15,316 ± 251 kg/ha) projected for 2050, assuming yield increases observed for the past 60 yr continue, would not be possible with projected increases in VPD, creating a water ceiling to maize yields. Substantial increases in maize yields and the production of high yielding grasses for bioenergy will require developing cultivars with greater water use efficiency, a trait that has not been a priority for breeders in the past.