Dynamic Adjustment in US Agriculture under Climate Change

We construct a stochastic dynamic dual model to investigate the structural adjustment of two aggregate output and three aggregate input categories in US agriculture under stochastic climatic change. More than a century of national annual data (1910-2011) is used in the empirical analysis. No constraints on asset fixity are imposed. Results indicate that, with rational expectations, both output categories as well as all input categories exhibit quasi-fixity in response to market change and stochastic climate change. Crops adjust more than twice as fast as livestock-49% versus 20% of the way toward their long-run equilibrium in one year. Fertilizer adjusts most rapidly toward equilibrium levels (88% in one year), and capital adjusts most slowly (5% in one year). Labor oscillates rather than converging smoothly toward equilibrium; its distance from equilibrium is the same as if it adjusted 59% of the way toward its optimal level in one year. Failing to anticipate climate change dramatically slows the estimated rate of adjustment for two netputs and modestly speeds the rate for two others, thus likely increasing overall adjustment costs. Failing to account for uncertainty in anticipated climate change has little impact on adjustment rates.