[I wish I had posted this on May 7* (see background post, one that I wish I had written on April 26).]
As part of Copenhagen Consensus 2012, I wrote a "perspective" paper [pdf] on the biodiversity challenge paper* along with my graduate school office mate and biodiversity expert Paul Chambers (the paper would have been technical drivel without Paul). Perspective papers:
... balance the challenge paper and indicate important issues that were not sufficiently dealt with in the challenge paper. The perspective papers are short papers, where published research that might have been left out in the original challenge paper will be reviewed, and alternate interpretations provided on the estimates and/or other strengths, weaknesses and omissions in the economic models pointed out. Their role is primarily to spur discussion and reveal substantial professional differences regarding the subject.
Here is a bit of Bjorn Lomborg's summary of the biodiversity paper:
The issue of disappearing biodiversity has increasingly received mainstream media attention in the past few years, and is starting to compete with climate change as the environmental threat that we talk about the most. Often, biodiversity campaigners have attempted to capture our attention with pictures of cuddly endangered animals or alarming figures about the rate of disappearing species. ...
In a research paper on biodiversity released today for Copenhagen Consensus 2012, Salman Hussain and Anil Markandya find that there will be a significant loss of biodiversity over the next 40 years. ...
The first solution focuses on increasing agricultural productivity through research and development. This may seem like a roundabout way to address biodiversity, but as the global population has increased to 7 billion, we have cut down more and more forest to grow our food. Between now and 2050, we will likely expand agricultural area another 10 percent, and that land will come from forests and grasslands. Thus, if we could increase agricultural productivity, we would need to take less and be able to leave more to nature. ...The authors estimate that with a $14.5 billion annual infusion into research, we can achieve 20 percent higher annual growth rates for crops and 40 percent higher growth rates for livestock, which over the next 40 years will significantly reduce the pressures on nature. ...
When we take into account that these forests will store more carbon, for every dollar spent, we will do about seven times the amount of good both for biodiversity and climate. ...Hussain and Markandya ... explore increasing protected land to about 20 percent globally (across a large number of ecological regions), over three decades. There are obvious benefits but also significant costs, principally the loss of output from the land that is taken out of use. ...
Land scarcity arising from such a policy would likely force an increase in agricultural productivity. The cost estimates for the newly protected lands have a big impact on the overall results. With higher assumptions, the program costs more than it achieves, even when the benefits of avoided climate change are included. With lower assumptions it only barely passes, making $1 spent achieve slightly more than a $1 worth of good. ...
The final program Hussain and Markandya propose seeks to prevent all dense forests from being converted to agriculture over a 30-year period. The academics do not attempt to assess the political viability of such an approach. To use the same measure as above, it would save more than seven times the area of Spain in tropical forests.
The benefits are very high, but there is considerable uncertainty about the costs. With estimates they find reasonable, the benefits exceed the costs even without including the CO2 storage value, and the solution is attractive because it will get a minimum of $7 back on the dollar.
In an otherwise remarkable paper, Hussain and Markandya do not develop confidence intervals around their benefit estimates nor conduct sensitivity analysis over assumptions use to develop the benefit estimates. In our perspective paper we reviewed a number of reasons for why the biodiversity benefit estimates may be more uncertain than the cost estimates. From our presentation** [PPT]:
We conducted an ex-post Monte Carlo sensitivity analysis in order to:
- provide a mean benefit-cost ratio for each policy over the range of cost estimates
- assess the sensitivity of the mean benefit-cost ratios to assumptions made in the analysis.
Our results are suggestive:
- none of the policies are likely to have benefits greater than costs at the 95% level of confidence when the standard deviation is about 40% of the point estimate of the present value of benefits.
Next post: CC12 Outcome.
*Note: I had hoped to stay on top of this but my blogging had to take a backseat to directing and grading 13 senior seminar papers. Graduation was yesterday so today I'm trying to fully understand what all I put on the backburner over the past month.
**Note: Alas, there was not sufficient funding for us to deliver our presentation in person.