@article{Bullock:288635,
      recid = {288635},
      author = {Bullock, David W.},
      title = {The Influence of State-Level Production Outcomes Upon U.S.  National Corn and Soybean Production: A Novel Application  of Correlated Component Regression},
      address = {2019-05-22},
      number = {1187-2019-1851},
      series = {Agribusiness and Applied Economics Report No. 790},
      month = {May},
      year = {2019},
      abstract = {Over the past 20 years, U.S. agriculture has witnessed  profound changes with respect to technology, climate, farm  policy, and other factors (ethanol production, Chinese  demand, etc.) that have major repercussions with regards to  the geographic distribution of crop production —  particularly, from a market share and geographic basis.  There have been many recent studies that have examined both  the direct and indirect impacts of these production factors  upon crop yields, acreage, and production from both a  temporal and spatial perspective. However, little to no  attention has been paid to the impact of these factors upon  the relative influence of each individual state’s crop  production outcomes as they relate to the national outcome.  The purpose of this study is to address this question of  state-level geographic importance for U.S. corn and  soybeans by employing the following procedure. First, a  metric is constructed to measure crop production outcomes  at any geographic level by comparing the current year’s  production to the recent historical norm. This metric,  called a production performance index (PPI), is simply the  difference between the current year’s crop production and  the Olympic average (drop minimum and maximum and take  arithmetic average of remaining values) of the previous  five years of production. The dataset used in the study  includes annual crop production values for the 1970 through  the 2017 crop years. The PPI, given its five-year lag, is  calculated with values for the U.S., each major producing  state, and the “Other States” residual from the 1975 to  2017 crop years for both corn and soybeans. The PPI time  series is divided into two distinct sets of time periods as  a proxy for the changes mentioned above: (1) the 1975 to  1995 crop years, and (2) the 1996 to 2017 crop years. The  1996 crop year was chosen as the dividing point since it  represents a watershed year in U.S. corn and soybean  production — the commercialization of the first GMO corn  (Bt corn) and soybean (Roundup Ready) varieties. Each  states’ relative influence upon the national production  performance outcome is determined by regressing the  individual states’ PPI values upon the national PPI value  for corn and soybeans under each time period. The  regression analysis is conducted using correlated component  regression (CCR) – a relatively new statistical tool for  sparse and mulicollinear datasets. The absolute value of  the standardized coefficient values from the regression  model are used to rank each state with regards to its  influence. Each state’s percentage share of the sum of the  absolute coefficient values was also calculated and used to  calculate a Herfindahl-Hirschman Index (HHI) by summing the  squared values of the percentage shares. The HHI is used as  a measure of the geographic dispersion of production  importance for the national aggregate. Overall, the results  showed a shifting geographic dynamic for both corn and  soybeans with the emphasis shifting from east to west in  general direction. This makes intuitive sense as many of  the observed technological and climatic changes over the  past several decades point towards corn and soybean  varieties that require a shorter growing season, and the  increase in the number of frost-free days in many of the  states in the northern reaches of the U.S. Corn Belt  region. Additionally, the greater utilization of irrigation  in crop production has likely contributed to the westward  expansion of both corn and soybean production — often at  the expense of wheat and cotton production. The slight  decline in the HHI for corn indicates that production  influence is becoming slightly more diversified from a  geographic perspective. For soybeans, the opposite effect  has occurred with a slight increase in the HHI pointing  towards greater influence from the key producing states of  Iowa, Minnesota, and Illinois — likely the result of a  shift from corn to soybean acres as all three states lost  influence shares in corn production between the two time  periods. },
      url = {http://ageconsearch.umn.edu/record/288635},
      doi = {https://doi.org/10.22004/ag.econ.288635},
}