General discussion

A link to the Millennium Development Goals Dashboard website:

http://esl.jrc.it/dc/mdg_download.htm

Looking at the question posted for this session, I think it will be important that we distinguish between metrics for use in a particular management situation (context) and metrics at a macro and more conceptual level.  The need for such a distinction should not surprise us: in every day life we would look for different (and probably participatory) metrics for evaluating (say) a local proposal to sell off a piece of town owned land for development than we would for evaluating the likely impact on the national economy of health care reform.  

The goal, it seems to me, is not that these metrics for micro and macro evaluation be the same.  Rather, its that the project-specific work should be disciplined to address the core conceptual elements defined in in the more rigorous but abstract macro metrics system.  In our particular case, if we endorse the macro framework presented by Arrow, Dasgupta et al., this would mean working to assure that project specific indicators did indeed address all the elements of inclusive wealth... or say what the alternative conceptual framework is that they would prefer to work within.  What will not advance sustainability as a science grounded field is to insist that every local case gets to design its own metrics that it then calls sustainability metrics.  (This would be optimizing on process legitimacy and perhaps policy salience over technical credibility).

A dozen years ago, when the National Academy study Our common future addressed the metrics question, we found that this ad-hocery is exactly what was going on.  Scores of metrics existed, none of which had developed a coherent conceptual foundation and none of which therefore had any defensible claim to superiority.  I hope the field -- or at least our discussions -- can move beyond that unproductive and unsatisfactory situation.

The Arrow et al. article and accompanying chapter expand the earlier Dasgupta chapter to explain how the maximizing wealth function could be used to create metrics for sustainable development. Assuming data is available on the true value of human capital, natural capital and other sticky areas, then this could be a standard way to measure progress towards sustainable development. I address these ideas below, and do not address project level metrics for SD.

Market pricing

I do not claim to understand all of the concepts underlying the Arrow et al. approach. However, I am somewhat skeptical that markets provide true, revealed value through prices for goods, including relatively simple capital goods. In the past two years, food prices have skyrocketed, in part due to speculation in financial markets justified on the grounds of liquidity and in part due to crops being used for energy (and a host of other reasons). Then, we witnessed a financial crisis, where the market rapidly devalued goods and services (e.g. US housing stock prices). What did we know about the value of these goods through these market price changes? How would we account for these changes using a sustainable development model that tracks progress over time?

Equity

There are also equity concerns we have discussed in the seminar, but that are not drawn out in the text. Was the grain going towards energy better served than the grain going towards food, because there was a higher willingness to pay? If this rearrangement improved overall wealth (even considering diminishing marginal returns from wealth), would it be preferred? These questions begin to get into the challenges of equity and “who’s sustainable development,” which we have touched on throughout the semester. However, as the Arrow paper reads, it seems that simply having markets would allow us to create these metrics through revealed prices; I don’t think valuation is quite that simple.

Incommensurability

In addition, perhaps this is thought to be built into the pricing itself, but the issue of incommensurability is not addressed. How can we compare the value of, for example an undergraduate education with a barrel of oil extracted? In one instance, the human capital investment is made under uncertainty – how long will this person live? What will their earning potential be given this investment? While we could estimate given current averages, this is unlikely to reflect the actual value of that person’s education. In addition, some goods may have aspects that are not fully captured in the price, for example spiritual values or intrinsic values. Finally, I do not understand how the model compares a depletable resource with generational gains in one time period; this seems difficult. What if the generational gains affect future generations (e.g. a mother receiving education, not contracting HIV and raising her own daughter in relative wealth.) Are some of these problems embedded or complex?

Describing limitations of the framework using other academic disciplines

I agree that this kind of framework (which includes measuring the stocks of resources and not just the flows, considering carbon emission, etc.) is no doubt an improvement upon current economic practice. And if the data were available to improve the system, by pricing natural capital (e.g. through REDD or other systems), then it would no doubt be beneficial in advancing sustainability science goals. However, it is important to also talk about the shortcomings of this approach, particularly in such an interdisciplinary book. The Arrow conclusions largely point to data limitations and cross country comparison limitations; in the actual chapter, it may be useful to bring in perspectives from other disciplines to provide a fulsome critique.

Institutions/policies and prices

Finally, one thing I was not sure after reading the chapter, was the role for institutions. Policies can create markets and prices for goods, which can be above or below the actual “value” of that good, as calculated with an economic model (e.g. feed-in-tariffs, carbon prices, etc.). Would this overall sustainability metric then judge the policy as positive or negative towards advancing SD efficiently? If so, the model itself would have to be quite objective and scientific in approach (which I’m sure it would be); nevertheless, it is probably impossible for the metric to be entirely unbiased or without value-laden assumptions. How would this affect policy evaluation? I am not clear on this after reading the article.

2) I found the comment that the statistical value of a life measurement is controversial because it has been found to be lower in poor countries very interesting. Clearly any idea that values one life above another could perhaps should be seen as morally controversial or wrong. However, this is a values question not an economics question. The methodology for getting a statistical life value seems reasonable and in fact from my experience people seem much more willing to take risks in developing countries with their lives* (maybe willing is not the right way to think about it as many of them have no other choice). But in general the measurement seems accurate to empirics or what we find on the ground, the question then becomes a moral/values one: Is this the way we think the world should be or do we want to change this reality and create a more equitable world. In this case is it better to include the reality or the ideal in our model for well-being?

Comments to Session 12 by CIECO-UNAM Group.

Topic 1. Measurements

In relation to the variables involved in the model presented for the second Task “Making definition operational”

a.       We wonder how much all the variables relevant for such a model for evaluating sustainable development are operational, that is they are measurable. If they are not measurable, or cannot be so at present, then they will be discarded from the model. Probably, it is important to make clear what does “operational” mean for the understanding of the operational definition and the importance of the associated measures.

b.      The model assumes that all variables are continuous. Yet, there are some qualitative variables, which would be relevant in terms of sustainability, that are expressed as discrete values or categories, such as access to health services or access to water (yes/no). Theses discrete variables would not be easy to incorporate into a differential equation. A different kind of model might be needed.  

Topic 2. Aggregation

We are concerned about the difficulties and limitations exposed for task 2 (making definition operational) and task 3 (empirical measures) such as implicit assumptions, data gaps generated and so on that results in a limited ability of an aggregate index to measure sustainability. Given this we agree that is an actual challenge the development of more efficient metrics for this purpose.

Topic 3. Use of Sustainability metrics

The inclusion of all environmental services or stocks of capital in the global market and assign them monetary values implies that these elements are interchangeable, transferable. Yet, this is not the case of oxygen, water or biodiversity.

A definition of sustainable development talks about intergenerational well-being, but, do we know the demands of the future generations? Are they the same as we have?

In the context of the metrics of sustainable development it would be important to consider a scale more wide than just regional units, given the fact that the regional economies are not isolated and neither are natural support systems.

I wanted to follow-up briefly on the discussion the cambridge group had at the end of class. We were discussing the tension between the usefulness of the comprehensive well-being idea as a conceptual framework that ensures every aspect of human well-being is accounted for in our projection of sustainability and the application of the equations as an actual tool for calculations, where many members of our group still felt significant hesitation about its empirical results. In particular we were surprised that in the Arrow et al paper, China and the US were indicated to be on a sustainable pathway between the 70s and the 90s. This seemed curious to say the least to many in the group---environmentalist for as long as I can remember have been sounding a warning bell that the whole world can't have the ecological footprint of the average US citizen without destroying the planet. It seems strange to turn around and declare the US to be increasing in human well-being over this time period (Perhaps some of us were misunderstanding the paper, but then further clarification is needed).