Climate Econometrics
The Climate Econometrics (CE)[5] project at Nuffield (co-directed by Hendry and Pretis) brings together a multi-disciplinary group of researchers from economics, econometrics, computing, climate science, political science and geography.
The aim is to develop and apply econometric tools to empirical modelling in order to better understand both how humanity has affected the global climate and how humanity has been affected in turn. Econometrics has proved a useful toolkit for statistically modelling high-dimensional dynamic economic systems subject to wide-sense non-stationarity (from stochastic trends and location shifts), outliers, potential non-linearities and simultaneous interactions, based on relevant but incomplete economic theory, so requiring model selection. Figure 1.2 illustrates the extreme wide-sense non- stationarity and huge outliers for UK domestic carbon dioxide (CO2) emissions in tons per person per year from 1860 to 2017, also showing dramatic recent reductions following the Climate Change Act of 2008, with emissions now below levels recorded in the 1860s when the UK was the “workshop of the world”.
1875 1900 1925 1950 1975 2000
Figure 1.2 UK domestic CO2 emissions in tons per person per year, 1860-2018
Since climate change is driven by economic activity creating greenhouse gas emissions (primarily CO2), similar econometrics modelling tools are proving valuable. Castle and Hendry (2019) and Castle et al. (2019) provide nontechnical explanations of the modelling tools and forecasting methods respectively. Empirical examples include linking econometric models and climate systems (see Pretis 2020); damage caused by hurricanes (see Martinez https://sites.google.com/view/andrewbmartinez/current-research/damage- prediction-tool); modelling data on UK and global CO2 emissions (see Figure 1.2 for the former and Hendry and Pretis 2013 for the latter); the impact of volcanic eruptions on temperatures (see Pretis et al.
2016) and of temperature rises on output worldwide (see Pretis et al. 2018b); and even the role of CO2 during past ice ages (Castle and Hendry 2020).Post-doctoral researchers on the climate team include Doornik (econometrics and computing: see Doornik 2008), Luke Jackson (oceanography: see Jackson and Jevrejeva 2016), Ryan Rafaty (climate policy: see Farmer et al. 2019), Sam Rowan (climate policy: see Rowan 2019), with Susana Martins (financial econometrics) and Jiao (econometrics) just joining, as well as DPhil students Moritz Schwarz and Jonas Krule, and research assistant Lisa Thalheimer. Angela Wenham is the communications officer and DPhil graduate Martinez recently left (econometrics and forecasting: see Castle et al. 2017).
Such a project also reflects a number of general developments in econometrics which have led from single topics to multi-disciplinary studies, from single authors to multiple, and spreading from being primarily economics focused to seeing applications in many other observational-data disciplines. An interesting future for Oxford econometrics lies ahead.