Past Energy Transitions – Day 3 recap

Chuck Mason, University of Wyoming, started off the day with his keynote talk on the risks of transporting crude oil by rail vs. pipeline. The boom started by fracking in the US has shifted the locations of production away from previous trends and to both new fields (e.g. North Dakota) and back to some of the oldest production fields. This has led to increased transportation by rail as well as increased interest in new pipeline investments. The risks vary. Spectacular accidents make it challenging to compare these risks and to determine the best transportation investments.

It turns out that the combination of serious and minor incidents by rail follows predictable patterns. First, there’s a negative relationship between accumulated minor incidents and the time between serious events, and then there is a positive relationship between rail traffic levels and the probability distribution over minor incidents. In other words, the more rail cars, the more accidents per week and the more minor incidents, the more serious accidents.

This then is related to damage costs. For additional rail traffic, expected costs from lost product, damaged capital, and some clean up costs go up. For every additional 1000 rail cars, we can expect $725 more in damages. With growth in cars carrying oil increasing at a rapid (though variable) rate over the last decade, the cost of rail accidents is adding up.

This research also provides important insights into how we must think about risks and damages in comparing long run investment choices relating to transportation of energy resources. While spectacular, accidents like those pictured above should not drive knee-jerk regulatory responses. Now that we can understand how the risks from rail increase with rail traffic in a systematic relationship, we have a better handle on the margins of interest needed to compare to other transportation options of trucking and/or pipelines, and can make more beneficial decisions for society.

We then moved on to a session on Fuels and their purposes. Paul Sharp, SDU, started off the session with “A Micro Study of the Role of Energy for Economic Development in Denmark.” This provided an overview of new research to come, which will be an extension of his recent published work with Sofia Henriques, Lund University, in  The_Economic_History_Review 2016: 69(3). In this work, they show how cheap imported coal and feed from Britain was a sufficient substitute when Danish energy sources were insufficient to fuel the growth in agricultural development that the cooperative movements spawned. The new work presents the counter-factual of how Danish development might have looked without access to cheap coal. They are using WWI’s restricted access to provide a natural experiment. The expectation is slowed growth, with spatial differentiation reflecting local access to alternative energy sources.

The next presentation was made by PhD student Hana Nielsen, Lund University. She presented joint work with Sofia Henriques (Lund), Paul Warde (Cambridge) and Astrid Kander (Lund) on “Energy intensity and the productivity race in industry (1870-1935). In this work the author’s aim to add energy intensity into studies of manufacturing processes that have traditionally been dominated by labor intensity arguments. The patterns of energy intensity and labor intensity are shown to be quite different for several industries. The authors look to cross-country differences as well, to see if there are notable differences between the core developed countries (England, Germany, Czech Republic, USA) of the period and the peripheral, catch-up economies (Portugal, Denmark, Sweden) – in the end, where the difference lies is mainly coal-rich v coal-poor economies.

Energy productivity has increased much more slowly than labor productivity, as measured at the margins. Technical change must have an important role – and as the audience pointed out – so must complementarities between energy and labor that are hard to quantify.  Using data from the countries above, three categories of manufacturing goods (food production, consumer goods, and capital goods), and three snapshots of data (1870, 1913, and 1935), the researchers generate a measure of energy consumption per ton of production. They find a constantly increasing energy/labor ratio over the time period. They attribute changes in energy productivity to increases in thermal efficiency and decreases in energy losses.  Their findings that much of the technological change was energy using but labor augmenting in this period concurs with other research in the later half of the 20th century.

The final morning presentation came from Harilaos Kitsikopoulos (Prometheus Unbound) who presented on “The Cost of Alternative Power Sources – A Comparison.” In this project, Dr. Kitsikopoulos has created impressive accounts for capital and operating costs for water, horse, wind and steam (coal) power (Newcomen engines) in a comparable framework for the UK by county. These accounts allow comparisons of the relative costs of steam, water and horses in HP in a complete way that elaborates the incentives to use (and/or shift) from one power source to another. It was not just the availability of a cheap power source that mattered; infrastructure costs for hydro, or high local costs for acquiring coal, might mean that lower efficiency was preferred due to cheaper costs.  This work provides fascinating and useful insights into the roles of cost in energy transitions.

In the afternoon, Sofia Henriques presented “From Primary to Useful En(X)ergy: Denmark 1870-2013.  This is part of a larger project on Energy Use and Economic Grwoth in Europe: 1870-2013 with Astrid Kander (Lund), Paul Sharp (SDU), David Stern (ANU) and Akshay Shanker (ANU), financed by the Swedish Handelsbanken.

The work is trying to find better indicators for observing the relationship between energy and economic growth than primary energy use – are there energy quality indicators that can better explain the evolution of energy efficiency and that can be used in growth production functions to test the role of energy in creating economic growth?

The stakes are high: decoupling energy intensity and carbon emissions is key to long run growth without significant climate impacts. The perils stem in part from conflation of technological change and energy efficiency gains – energy intensity changes get used as measures for energy efficiency, but this is a poor proxy. There is a desire to bring in more engineering measures of energy quality (exergy analysis) to capture the potential to do work, or the available energy for work. These are complicated calculations, but if sorted through, can improve understanding of energy transitions in the past and yet to come by converting primary energy data into available energy data.

The final presentation was mine; I won’t spend a long time on discussing it here. I’m looking at the transition from sperm whale oil to kerosene for lighting purposes in the mid 19th Century. The whaling and early petroleum industries were significant components of the American economy of the 19th Century, and each has been studied at length. But though much is made of the fact that kerosene quickly outcompeted sperm whale oil, the two have never been studied together as a transition from one lighting source to another in a bio-economic framework. In taking this approach, my work aims to answer questions about how shifting costs and benefits of the whaling industry promoted the search for substitutes, and how the acquisition of these substitutes fed back into the whaling industry (and the conservation of the whales).

We concluded the meeting with discussions over general themes in past energy transitions, including the pace and impact of change as seen from contemporary and historical viewpoints, and the role of endowments and convertibility from primary energy to outputs as demands changed. We agreed there is a rich field of past energy transitions awaiting closer scrutiny and many useful lessons to be applied to the resource transitions underway today. So get to it!