Future Yield Growth: What Evidence from Historical Data?

Joint Program Report
Future Yield Growth: What Evidence from Historical Data?
Gitiaux, X., J. Reilly and S. Paltsev (2011)
Joint Program Report Series, 26 pages

Report 199 [Download]

Abstract/Summary:

The potential future role of biofuels has become an important topic in energy legislation as it is seen as a potential low carbon alternative to conventional fuels. Hence, future yield growth is an important topic from many perspectives, and given the extensions of the period over which data are available a re-evaluation of yields trends is in order. Our approach is to focus on time series analysis, and to improve upon past work by investigating yields of many major crops in many parts of the world. We also apply time series techniques that allow us to test for the persistence of a plateau pattern that has worried analysts, and that provide a better estimate of forecast uncertainty. The general conclusion from this time series analysis of yields is that casual observation or simple linear regression can lead to overconfidence in projections because of the failure to consider the likelihood of structural breaks.

Citation:

Gitiaux, X., J. Reilly and S. Paltsev (2011): Future Yield Growth: What Evidence from Historical Data?. Joint Program Report Series Report 199, 26 pages (http://globalchange.mit.edu/publication/14009)
  • Joint Program Report
Future Yield Growth: What Evidence from Historical Data?

Gitiaux, X., J. Reilly and S. Paltsev

Report 

199
26 pages
2011

Abstract/Summary: 

The potential future role of biofuels has become an important topic in energy legislation as it is seen as a potential low carbon alternative to conventional fuels. Hence, future yield growth is an important topic from many perspectives, and given the extensions of the period over which data are available a re-evaluation of yields trends is in order. Our approach is to focus on time series analysis, and to improve upon past work by investigating yields of many major crops in many parts of the world. We also apply time series techniques that allow us to test for the persistence of a plateau pattern that has worried analysts, and that provide a better estimate of forecast uncertainty. The general conclusion from this time series analysis of yields is that casual observation or simple linear regression can lead to overconfidence in projections because of the failure to consider the likelihood of structural breaks.