21 December 2005

Paper finds mode shifts more cost effective than technofixes for climate change mitigation

I have just digested a recent paper:
Lloyd Wright and Lewis Fulton (2005) “Climate Change Mitigation and Transport in Developing Nations,” Transport Reviews 25, 6, 691–717 (click here for journal contents but full access is for subscribers only; or peek here for a possibly-naughty pdf of the full paper)
I found it is a striking read. Assuming its methods can stand up to scrutiny then governments and funders of climate change mitigation projects should take notice... Are you listening GEF and the others?

It makes a strong case that an emphasis on encouraging mode shifts away from private vehicles (or even just slowing mode shifts towards private vehicles) is by far the most cost effective way to mitigate greenhouse emissions. In fact, fuel focused approaches are simply not cost-effective. Yet these are the projects that are getting the lion's share of funding!

Allow me to quote from its conclusions, starting on page 715:

... The scenario analyses indicated that the cost of fuel-based solutions ranged from approximately US$148 to over US$3500/tonne of CO2. By contrast, shifting mode share from high-emitting sources (private vehicles) to lower-emitting sources (public transport and nonmotorized options) produced emission reduction costs between US$14 and US$66/tonne of CO2.

This research has thus indicated that fuel-based solutions alone will not likely achieve cost-effective reductions in greenhouse gas emissions. The most cost effective means to emission reductions appears to be a diverse and integrated package of measures that promote shifts to lower-emitting modes.

Despite the transport sector representing the fastest growing source of greenhouse gas emissions worldwide, there has been relatively little project activity to address emissions from the sector. The number of transport projects under the mechanisms of the Kyoto Protocol and under the GEF is relatively small in comparison with other sectors.

While the projections of increased motorization indicated in the IEA reference case are a cause for concern, these trends are not preordained. An alternative is still achievable for most developing nation cities. The low-cost solutions that have emphasized public transport, bicycling and walking, and land-use changes in Bogota and Curitiba are certainly possible elsewhere. Whether the political will exists elsewhere is a question to be answered.

Other points in the paper that struck me:

  • the world is fast approaching 1 billion motor vehicles on its roads!
  • citing a study by the UNFCCC, Finland stands out as the only country to have decoupled transport sector growth from economic growth (with both the 'Nokia effect' in which a key economic driver is not transport intensive, and as a result of strong efforts to curb vehicle ownership and usage
The authors suggest a few plausible reasons for the popularity of technical fixes over efforts at modeshifting (quoting here from p. 715):
  • Technological solutions (tailpipe technologies, fuels, propulsion systems) can appear to be simple black-box solutions that are intrinsically easier for public officials to understand than a broader systems approach.
  • Higher-technology options may be perceived as being ‘modern’ by many political officials, while non-motorized transport may be perceived as counter to national aspirations.
  • It may be far more politically expedient to promote increased motorization rather than public transport and non-motorized transport.
  • BRT is a relatively new concept and there may be informational barriers to its wider application.
  • It is possible that simply improving the state of developing-nation footpaths could be one of the most effective long-term measures, from the perspectives of both cost and overall development. However, it is unlikely that any global footpaths initiative is on the horizon anytime soon.


Anonymous said...

I was impressed with the broad range of considerations in this study. It is nice to see land use and policy changes addressed in conjunction with other recommendations. The BRT boosterism seemed a little out of place and detracted from the effectiveness of the main argument that we should focus more on shifting mode share away from private auto use. Isn’t there much more extensive evidence to support this claim with traditional bus service? And is he implicitly arguing against other mass transit solutions like subways?

Anyway, it does seem from his statistics that investment in fuel technology is misguided. But I found a couple confusing points in the data. Maybe someone here can clarify? Even better if Mr. Wright reads this blog and can answer himself.

The first problem that I see is in table 5. The CO2/day is pretty much 5% of the CO2 over 20 years. This would mean that the 20 year figure is actually a 20 day figure. Also, in trying to check a couple calculations, I can’t get his figures to work out. It seems that CO2 tonnes/day should equal distance traveled/day * fuel liters/km * CO2 kg/liter divided by 1000 kg/tonne. However, I’m getting 3484 tonnes/day with this calculation instead of his figure of 1087.

Secondly, and maybe I’m reading this wrong, in table 6 Wright claims that the 21,744 tons of CO2 produced by autos (in 20 days or 20 years) is the result of their 20% share of trips. But I think it is actually the result of their 50% share of distance traveled (13, 333 km of a total 25,000 km/day for all modes), which would cut the reduction per cost in half.


Lloyd Wright said...

Dear Scott,

Many thanks for your feedback and astute review of the tables.

You are quite correct that there is a typo in Table 5 of the document. The eighth column should read “Tonnes of CO2 per year” and not “Tonnes of CO2 per day”. Fortunately, the next column, regarding the 20-year accumulation of CO2 savings is correctly labelled. I very much apologise for the error.

With regard to the actual calculation, I am attaching the actual spreadsheet utilised. This shold hopefully answer your question.

You are also correct that the 20% mode share for private vehicles based on the number of passenger-trips is equivalent to a 50% mode share based on the number of kilometres travelled. We have utilised the passenger-trip value as the determinant for “mode share” since this tends to be the most common convention. But we realise that one could also define mode share based on kilometres-travelled.

We tried very hard to not appear overtly biased to BRT. And in fact, we ran scenarios with rail-based technologies as well. Unfortunately, the results in terms of cost per tonne of CO2 reduced were several orders of magnitude beyond market prices. This was the case even assuming a largely hydro-based electricity supply. This is not to say that rail (metros and LRT) are not justifiable for other reasons. It is just a bit difficult to make the case based on CO2 offsets. BRT just happened to be the only mass transit option that produced positive results in terms of CO2 cost effectiveness. You are correct, though, that if one could achieve similar mode shifting results with conventional bus services, then this would also produce very cost-effective CO2 reductions.

Again, I apologise for the typo in Table 5 of the article. I am most appreciative that you have called this to my attention.

By the way, much of the analysis was made possible by the exhaustive work done by the IEA in collecting base data for CO2 calculations. I highly recommend downloading the IEA spreadsheet:


It really is a goldmine of information for a range of analytical work.

Best regards,

Lloyd Wright

Anonymous said...

I reviewed the tables and was able to match Lloyd's numbers with the correction of tonnes/day to tonnes/year.

Many thanks to Lloyd for providing the data and calculations. The IEA spreadshhet contains an overwhelming amount of data. I'm still trying to digest all of it.

While I understand that modeshare was represented by number of trips to follow convention, I still think it suggsts an unrealistic savings. If car trips are generally longer than BRT trips, then they cannot fully be replaced by BRT trips.

Thanks, Lloyd, for helping me clear up my questions.