Soil erosion is a threat to agriculture's capacity to feed the world's population. The objective of this study is to examine the ability for N and P fertilizers and precipitation to compensate for soil erosion. This study addresses the impact of soil erosion without the masked effects of improved technology over time. The effects of fertilizer application, precipitation, and soil removal on spring wheat (Triticum aestivum L.) yields were determined from an 8-yr experiment of Williams loam (fine-loamy, mixed, typic Argiborolls) near Sidney, MT. A quadratic production function is estimated using the experimental data, which contains observations on the effects of four erosion levels, and three N and three P application levels. The results indicate that every inch of soil loss decreases yield by an average of 0.75 bu/acre and that increased fertilizer applications can substitute for only small amounts of soil loss. Precipitation offers the best substitute for erosion, in that small increases in precipitation in this region offset the productivity impact of a wide range of erosion. The cost of supplementing precipitation as a means of offsetting erosion can be high, however. Erosion control strategies may be more cost effective.