This study has three main objectives: (i) to quantify the impacts of milk and feed price changes on the primary milk supply in the U.S.; (ii) to examine the impacts of technological changes on the price responsiveness of supply and specific herd characteristics; and (iii) to generate dynamic long-run forecasts of the milk supply response to price changes and possible future technological advancements. The econometric analysis contained in this study is an update of the model by Chavas and Klemme (1986). We used the residualbased bootstrap to test hypotheses regarding the long-run price-responsiveness of supply, and found that the 10-year elasticity of milk supply to milk price is lower in 2007 than it was in 1980. This result is most surprising. One might expect that with better genetics, improved heifer management and larger farms the industry would be likely to react to prices more quickly than almost thirty years ago, when small and medium-sized dairy operations played a major role. A detailed analysis of the predicted herd structure supports the conjecture that a decrease in price responsiveness is a consequence of decades-long excessive focus on yield improvement in genetic selection. The intensive production process could make cows susceptible to health problems, imposing biological constraints on the economic lifetime of a cow. Hence, herd expansion decisions will be harder to implement, as culling rates are not easily reduced, and more replacement heifers are needed just to keep the herd size stable.