This is an extract of Dr Jennifer L. Castle's keynote presentation to the 26th Dynamic Econometrics Conference at the Oxford Martin School, University of Oxford, on 5 April 2024, entitled 'The founding of a discipline: Sir David F. Hendry's contributions to Climate Econometrics', recognising Sir David Hendry's 80th Birthday.

The origins of Climate Econometrics

Climate Econometrics started life as 'Economic Modelling in a Rapidly Changing World' (EMoD); a programme aimed at developing new economic modelling techniques, funded by James Martin (Oxford Martin School) & George Soros (Open Society Foundations & INET Oxford). The Team, led by Sir David Hendry included many academics, post-docs & students, based at Eagle House in Jericho, Oxford.

The programme was designed to investigate changes needed to economic analyses, policy, empirical modelling and forecasting when sudden, or very rapid, unanticipated changes occur. It evolved out of the financial crisis: many factors were to blame for this, but inadequate economic models leading to a failure to forecast the crash was partly at fault, exacerbated by poor policy responses. This precipitated the need for a paradigm shift. The EMoD programme looked at developing new methodology and automatic modelling software to implement new approaches.

Climate Change, the most pressing issue facing humanity, was a major motivation for the evolution of EMoD to Climate Econometrics. Sir David Hendry, with a longstanding interest in the natural sciences, palaeontology & geology, was concerned about anthropogenic influences in climate change and the economic consequences thereof. At the time climate science tended to be non-stochastic and scenario-driven and David saw a role for econometrics in advancing our understanding and driving policy. Data speaks a thousand words and Climate Change is a natural application of the methodology and software developed by EMoD.

In 2015 a Robertson Grant led to the establishment of a 'Climate Econometrics' programme, now based in Nuffield College, with the aim of developing approaches appropriate to a world undergoing rapid unanticipated changes, and to improve forecasting methods in such a setting.

What is Climate Econometrics?

Climate Econometrics applies statistical methods to help disentangle complex relationships between human actions and climate responses and their associated economic effects. The research is two-fold: developing fundamental statistical theory and econometric methodology to support empirical analysis; and developing climate applications using innovative modelling techniques. Explaining empirical evidence requires taking account of climate science and human behaviour. What is observed is generated by an underlying data generating process (DGP) that is unknown and changing, so we must search for a close approximation to its evolution.

Climate Econometrics is both a research group producing academic papers, policy briefs and reports; and a community of researchers interested in econometric models of climate change. The network numbers over 200 members globally, with an annual conference on Econometric Models of Climate Change. We have an advisory board consisting of experts across the disciplines, which includes two Nobel prize winners. We run a monthly seminar series on Climate Econometrics (past seminars are available to view on the Climate Econometrics website), there is a Climate Econometrics blog, a bi-annual newsletter and frequent updating of the website and social media output. There have been two themed special issues of the Journal of Econometrics including ‘Econometric Models of Climate Change’ (2020) and ‘Climate Econometrics’ (2024) - and we have supported two themed special issues in Energy Economics including ‘Econometrics of Climate, Energy, and Resources’ (2023) and ‘New Developments in the Econometric Analysis of Energy and Climate’ (2022).

The Climate Econometrics research group develops methodology to locate an unknown DGP nested within a large set of possible explanations, including dynamics, outliers, shifts, and non-linearities. They investigate taxonomies to understand forecast performance, where forecast errors are likely to arise, and how to mitigate such forecast errors, and they develop tools to enable policy analysis, including automatic tests for a lack of invariance of models to the policy changes they suggested.

Climate Economics research activities include:

• Climate modelling
• Climate Policy
• Sea level rise
• Forecasting facing breaks
• Hurricanes & volcanoes
• Modelling with shifts
• Model discovery
• Software implementation
• Health and Climate
• Just Transition

The Econometric Modelling of Climate Change Conference Series has become a well-established forum:

• 2016: 1st – University of Aarhus, Denmark
• 2017: 2nd – University of Oxford, UK
• 2018: 3rd – Universit ´a di Roma ‘Tor Vergata’
• 2019: 4th – University of Milan-Bicocca (with Claudio Morana)
• 2020: Postponed due to Covid19
• 2021: 5th – University of Victoria, British Columbia
• 2022: 6th – University of Toulouse
• 2023: 7th – Vrije Universiteit Amsterdam
• 2024: 8th – King’s College, University of Cambridge, August 16-17

Sir David Hendry's influence on the field

David's influence on the field has been immense, but here I pick four of my favourite contributions:

In Climate Change: Lessons for our Future from the Distant Past, Hendry outlines how past climate change has been driven by natural forces (plate tectonics, volcanism & orbital variations). Life has survived great changes and thrived in very different global temperatures, but en route huge numbers of species went extinct, even if long after, new species evolved for the new environment. Change is the key word—humanity is now changing the climate by vast emissions of greenhouse gases, mainly CO2 from burning fossil fuels. Economic analysis offers a number of ideas, but the key problem is that distributions can shift, making action to avoid possible future shifts urgent and the saturation technology developed by the group is crucial to modelling such shifts.

In Anthropogenic influences on atmospheric CO2, with Felix Pretis, Hendry looks at more recent evidence from Mauna Loa observatory, using Impulse Indicator Saturation and automatic model selection across a wide range of climatic and economic variables to determine the extent to which anthropogenic sources increase atmospheric CO2.

In First in, First out: Econometric Modelling of UK Annual CO2 Emissions, 1860–2017, Hendry developed a model of UK annual CO2 emissions per capita, which are now far below the levels in the 1860s.

In Five Sensitive Intervention Points to Achieve Climate Neutrality by 2050, Hendry takes identifies Sensitive Intervention Points to map a path to net zero These include:

1. Installing sufficient non-greenhouse gas electricity
2. Electric vehicles connected to a smart grid to increase storage capacity rapidly
3. Use any surplus energy for hydrogen production
4. The use of liquified hydrogen for industry
5. The use of electricity in agriculture

Why Sir David Hendry's climate work matters

Econometric modelling techniques really matter for understanding, modelling and forecasting climate change. To understand the very messy data we need to be able to filter that data and process it in a way that helps us to interpret the causal mechanisms driving climate change. Elegant theories are not sufficient: they need to be taken to the data and rigorously tested - this is hard to do when nothing is constant, and that is why the econometric developments underpinning David’s research are fundamental.

David’s vision with Climate Econometrics provided legitimacy to a new econometric field that empowered others to work on climate-econometric questions. New PhD students are keen to work on climate econometrics and the material is now being taught in masters courses. David’s work has been characterized by great friendships and collaborations and a great generosity in academic research and spirit. While climate change remains a major global challenge, progress can be made. Modern dynamic econometric analysis can shed light on climate change and help guide policy. To what do we owe the success of Climate Econometrics? Sir David Hendry's Vision, Passion & Thirst for Knowledge.

Dr Jennifer L. Castle is Director of the Climate Econometrics Programme at the University of Oxford.

The 26th Dynamic Econometrics conference was held in Oxford, UK on 3-5 April 2024. The conference provided a forum for presenting and exchanging research results and practical experience in empirical economics, time-series and cross-section econometrics, computational and financial econometrics, and applied mathematics. This year, the conference celebrated Prof. Sir David F. Hendry's 80th birthday and covered topics in which David, an IIF Fellow, has made pivotal contributions.