Asking the wrong question (more on ecological economics)

At the back of the environmental valuation chapter of the environmental econ text I'm using for the first time this semester, Goodstein uncritically writes:

In a very ambitious article in the journal Nature, an interdisciplinary group of scientists attempted to put a value on the sum total of global ecosystem services--ranging from waste treatment, to gas regulation, to nutrient cycling (Costanza et al. 1997). They concluded, stressing the huge uncertainties involved, that ecosystems provide a flow of market and nonmarket use value greater than $33 trillion per per, as against planetary product (world GDP) of $18 trillion.

I've written on how this estimate doesn't make sense since value can't be above income. One hangup with this criticism is the nature of the question. If the world exploded then, of course, we'd be missing out on more than world GDP. These aren't the questions where we need economic values.

The primary purpose of environmental valuation methods is to provide a benefit measure to compare against costs for benefit-cost analysis of environmental policy. The types of questions typically considered by benefit-cost analyses are marginal decisions (even climate change). In order words, what are the benefits and costs of an incremental improvement in environmental quality? The "what is the value of god?"-type questions don't make sense, at least not to me.

If you are interested in a better written, thought out, etc reply to the Costanza, et al paper (and others), check out "On Measuring Economic Values for Nature" by Bockstael, et al. Here is the abstract. Some excerpts:

Costanza et al. (9) and Pimentel et al. (11) calculate "total values" by scaling up a variety of estimates of values taken from other studies of ecosystem services and functions. The original studies valued small changes in specific and localized components of individual ecosystems, with each study implicitly holding constant other features of the global ecosystem and the economy. It is incorrect to extrapolate the value estimates obtained in any of these studies to a much larger scale, let alone to suppose that the extrapolated estimates could then be added together and applied to the whole planet.

/.../

... at least one aspect of the errors of scaling up values can be easily illustrated. Whatever the private good involved in consumption choices, be it ball point pens or pairs of socks, we can expect that it will be subject to diminishing value as more of the good is purchased (and consumed) within a given (conveniently defined) period of time. In applications, this property implicitly reflects the fact that even though not all pairs of socks are the same, in most cases an individual's willingness to pay for another pair is likely to diminish the more he already has. In this example, the individual has plenty of close substitutes for this potential new purchase. Likewise, an individual's willingness to pay for another pen will be conditioned on how many others he already has. Take away those previously purchased socks or pens, and the willingness to pay for a new one is very different.

Thus, simple multiplication of a physical quantity by "unit value" (derived from a case study that estimated the economic value for a specific resource) is a serious error. Small changes in an ecosystem's services do not adequately characterize, with simple multipliers, the loss of a global ecosystem. Values estimated at one scale cannot be expanded by a convenient physical index of area, such as hectares, to another scale; nor can two separate value estimates, derived under different contexts, simply be added together. When we estimate a compensation measure for one element of an ecosystem, we assume that other aspects of the world that influence human well-being are unchanged. For example, we might compute a compensation measure for the elimination of a specific wetland, holding others constant. In another analysis, a compensation measure for the elimination of a different wetland might have been estimated, holding the first at its initial level. But the two compensation measures cannot be added together to obtain the correct compensation in the event that both wetlands are eliminated.

/.../

Clearly, extrapolating from small-scale studies cannot help us to estimate the economic value for the world's ecosystems, but is there any reasonable answer to this question? We know that Costanza et al.'s estimate of an annual economic value of 33 trillion U.S. dollars is a logically inconsistent measure of what individuals would be willing to pay to avoid the loss, if only because this estimate exceeds their total ability to pay. Simply put, if, as Costanza et al. (9) estimate, the world's GNP is 18 trillion dollars, the world's population does not have 33 trillion dollars to spend annually.

/.../

Suppose instead that one asks what compensation the people of the world would require in order to voluntarily give up the world's ecosystems. This willingness-to-accept form of the economic valuation question uses the alternative baseline and highlights the need to define clearly the alternative state that applies when the compensation is paid. This is required to measure the compensation necessary to equate each individual's well-being, given the current level of the world's ecosystems, with that in an alternative state. But what is this alternative state: a different, newly emerging set of ecosystems or a complete void?

/.../

If the alternative state is "nothingness", then the answer to the willingness to accept question is trivial. There is no finite compensation that individuals would accept to agree to the loss of the world's ecosystems, and they would pay everything they had to avoid it. To an economist, this is the definition of an essential good, a good for which there is no finite compensation for its complete elimination. In this sense, ecosystems are essential.

/.../

This paper is not intended as a judgment on the appropriate definition of value. It is about furthering good science. Without cooperation, ecologists and economists cannot serve the interests of society. Moreover, cooperation is impossible without a better appreciation and respect for each others' discipline.

It seems that there are two sorts of ecological economists. The first sort is the environmental economist who grew frustrated with some of the constraints of neoclassical economics but did not reject economics. I think of these types as "economic ecological economists." The work they do is environmental economics with greater integration of ecology and maybe a bit more fire and brimstone. These can all be good things.

The second sort is the ecologist who grew frustrated with economics and rejects neoclassical economics. In fact, this type is happy to stand up at a conference and tell everyone what is wrong with economics and economists. I think of these types as "ecological ecological economists." The work they do is ecology with some economics (and a lot of fire and brimstone) thrown in, and a lot of times the economics isn't always exactly right.

For me, the bottom line is: beware environmental values developed by the ecological ecological economist.

As a reader, recognize another bottom line: economists don't like ecologists invading their turf (and vice versa). I'm awake enough to realize that I'm probably guilty of that, sure.

Comments

[allen claxton]

Goodstein uncritically writes:

insert quote

That Goodstein always was a sloppy bastard.

[John Whitehead]

Yikes!

[John Whitehead]

Note: The first two comments above refer to an unfortunate incident where the author of the post unfortunately posted it before it was completely written. Luckily, for me, all of the profanity had already been deleted! ;->

[Dano]

Well, I'm definitely going to download this paper, but whether I can get to it right away is something else altogether.

We know that Costanza et al.'s estimate of an annual economic value of 33 trillion U.S. dollars is a logically inconsistent measure of what individuals would be willing to pay to avoid the loss, if only because this estimate exceeds their total ability to pay.

This is not logically inconsistent.

The Costanza et al. paper says (I've met Robert and I've talked to him about this paper) that ecosystem services are not properly valued in the economy.

Certainly the harrumphing here: Thus, simple multiplication of a physical quantity by "unit value" (derived from a case study that estimated the economic value for a specific resource) is a serious error.

is likely true. So a first paper has a serious error - stop the presses! The point is to highlight the value of ecosystem services; in this regard, the paper has been wildly successful.

Suppose instead that one asks what compensation the people of the world would require in order to voluntarily give up the world's ecosystems. This willingness-to-accept form of the economic valuation question uses the alternative baseline and highlights the need to define clearly the alternative state that applies when the compensation is paid.

This is gibberish. The alternative state is no ecosystem services, which means no economy and no life.

There will be no willingness-to-accept, because there will be no one there to will anything.

Best,

D

[John Whitehead]

Dano,

The Bockstael paper makes the same points that you do.

They also point out that this is a useful exercise (the value of the world is big, the world is important) as long as this type of economic value is not used for policy analysis.

Unfortunately, the authors and other ecological ecological economists do advocate using this type of value for policy analyis.

Since this type of economic value is severely biased upwards, when plugged into a benefit-cost analysis, maximum environmental protection will always be preferred (which is the ultimate goal of this type of economic value).

This is the point I was trying to make here.

[Dano]

Sure, John, and looks like I said the same thing there too as here.

Anyway, I'm getting an ACS website temporarily unavailable, so I can't read the paper to properly speak to it.

Again, I'm arguing in a Daly-esque manner that ecosystem services aren't properly entrained in current economic systems. If that's what Bockstael et al. are saying too then I'll shut up now.

Best,

D

[Dave Iverson]

Come on John. Do you think that Kenneth Arrow and coauthors were playing the game that "maximum environmental protection will always be preferred (which is the ultimate goal of this type of economic value)" ?

I don't, but rather agree with Dano that "the point is to highlight the value of ecosystem services." As I've said before, I like the way Mark Sagoff and others work in the story of the worth of ecosystem services in qualitative, narrative form, much better than the way Costanza, Arrow and others did in the Science article. Still, I can understand their "fighting fire with fire" approach.

What I can't understand is why you choose to single out this particular "error" in a field so full or error. See, e.g. my Top Ten Reasons Why Cost Benefit Analysis Fails in Public Choice Settings!

Or maybe I do so understand, since you'd have to change your economic religion to approach the problem the way I do.

[John Whitehead]

Dave,

The authors on the Costanza et al paper are: Costanza, d'Arge, de Groot, Grasso, Hannon, Limburg, Naeem, O'Neill, Paruelo, Raskin, Sutton & van den Belt.

However, Arrow is a coauthor of this piece, placing him firmly in the BCA camp. Here is the conclusion:

Benefit-cost analysis can play an important role in legislative and regulatory policy debates on protecting and improving health, safety, and the natural environment. Although formal benefit-cost analysis should not be viewed as either necessary or sufficient for designing sensible public policy, it can provide an exceptionally useful framework for consistently organizing disparate information, and in this'way, it can greatly improve the process and, hence, the outcome of policy analysis. If propetly done, benefit-cost analysis can be of great help to agencies participating in the development of environmental, health, and safety regulations, and it can likewise be useful in evaluating agency decision-making and in shaping statutes.

I'll read your Top 10 ASAP.

[Dano]

3. CBA treats "consumers" as sovereign, and "citizens" as absent.

Even if we were to ignore communal responsibilities and think only about individuals, there is more to life than markets. All substantive human and/or environmental issues have roots at least as deep in who we are as in what we want.

Yes indeedy Dave.

As citizens, we value nearby nature, perhaps not explicitly and oftentimes not consciously (ergo not...ummmm...rationally), but there is increasing evidence that greenness has a calming effect on fatigue caused by directed attention.

Since we often don't consciously percieve this effect, we sometimes don't...ahem...rationally seek it out yet we unconsciously may do decision-making that priveleges greenness.

Conversely, we may not know of this calming effect of greenness on fatigue caused by directed attention, therefore making decisions that are...erm...not in our best interest.

As we don't have a way of valuating citizen issues in our economy ['Say, subject, what is your Willingness To Pay for engaging in a democratic dialogue on quality of life issues? How about your WTP for greening up your area? Is it more than a can of Pringles? A Prada bag?'] CBAs are only a start in understanding how to engage in and reflect upon ecosystem services. That is: economics is only a way to look at it if everyone knows equally and can calculate Pareto optima in a dialogue free of power relations.

Best,

D

[Dano]

Good quote John:

Although formal benefit-cost analysis should not be viewed as either necessary or sufficient for designing sensible public policy, it can provide an exceptionally useful framework for consistently organizing disparate information, and in this way, it can greatly improve the process and, hence, the outcome of policy analysis.

I have an abstract due at the end of the month on this very thing, where in the paper I argue that this view is sufficient only for a subset of the population being subject to policy-making. I'll read that paper.

I'll also calm down and save your bandwidth. :o)

Best,

D

[John Whitehead]

Dave,

At best, BCA is a tool that helps inform decision makers about the efficiency aspects of a policy issue. Decision makers should incorporate other parameters as they see fit. Economists are the decision makers, unless we get elected or appointed.

I think these are the things that your top 10 focus on. We agree that there is more to a good life and good public policy than efficiency.

Where we disagree is the role of the economist working as an economist. BCA is one of the things that we do best, and what we should bring to the table.

Other things are important, equity, etc, but I don't know anymore about these things than anyone else.

I'm happy if BCA helps avoid gross inefficiencies.

I'm unhappy when BCA uses bogus numbers to try to influence policy more than an efficiency analysis should warrant.

Is that a religious position? I don't think so.

Thanks,

John

I'm unaware of the alternative being proposed to BCA by those who do not consider it at all useful. I'm admittedly ignorant but would like to be enlightened as to the approach they suggest. Maybe a short version?

[allen claxton]

The second sort is the ecologist who grew frustrated with economics and rejects neoclassical economics.

Not that I really know, and I understand the point that you're making in presenting the two poles, but I think that there are useful gradations on the ecological ecological economics side of things.

At least, that's where I think you'd put an evolutionary economics approach (which seems very tangled up in much of ecological economics, though not necessarily the world-valuation schemes that get your ire). This focuses more on boundedly rational people, tacit knowledge, and path dependency and the like, and gets into "reject[ing] neoclassical economics," but more in a way of trying to improve economics' starting point than doing away with the enterprise as a whole.

[Dave Iverson]

John,

I guess I'll have to work my office a bit tomorrow to see if I can find the Science article that I thought had Arrow's name on along with Constanza. Perchance this aging brain of mine is playing tricks on me. I'll add more here then.

In the meantime, here is a link to an economic evaluation of these issues that I like, at least on initial inspection. Although I would like to quiz the authors a bit as to how one might conjur up their hypothetical 1 million dollars annually in lost environmetal services. It is titled Economic Pathways to Ecological Sustainability [PDF], Partha Dasgupta, Simon Levin, and Jane Lubchenco. I just found it while trying to find the other article online.

What I like about Dasgupta et al is the broad institutional framing of the issues. It's explanatory worth I find very helpful to policy makers. It helps policy makers better understand the complexities and nuances of proposed or possible action.

Like you, these authors dismiss the 33 trillion Costanza et al number as mostly meaningless. What they do not attempt is to evaluate the worth of a given project or program. Perchance they do so in other work. I don't know.

My angst over CBA comes in the main from thirty years of experience looking at inept CBA number-running by government economists. This does not mean that there aren't "very good" examples out there. Just that I find too many bad examples to be supportive of the practice.

Besides I can't imagine how one goes about providing useful information to decision makers except in the most trivial of cases.

Then there is the dilemma of how one arrives at social rather than private benefit. Then there it the problem of correct social rates of discount. Then there is....

Arrow's "if properly done" caveat is a big one, from my experience base. So I retreat to suggesting that people account for costs admiting the unreliability or future estimates, and project revenues where they are to be captured, again admiting the unknowability of future receipts. But I stop short of recommending closed-form CBA.

There is just too much that is unknown and/or unknowable. Besides, my work has been mainly in forestry where the time lags in production functions for growing trees and tracking biodiversity are big.

One final notion. I remember Ronald Coase arguing that things had to make sense in total AND at the margin. Just one more wrinkle in the CBA mess.

[Dave Iverson]

Someone above wanted to know about alternatives being proposed to BCA by those who do not consider it useful. There are no easy answers, but here are two book suggestions. Both are excellent:

Deborah Stone. Policy Paradox: The Art of Political Decision Making

James March. A primer on Decision Making: How Decisions Happen

[Robert Metcalfe]

Dave

I haven't read those books but I assume they advocate multi-criteria analysis (MCA)? This in itself is a flawed tool for policy since one cannot trade-off two competing interests easily and is utterly useless for efficient resource allocation. Furthermore, many other tools, such as MCA, rely on the preferences of the decision-makers rather than the preferences of individuals or consumers. However, given that we are trying to create a market for the public good, one needs to try to replicate a market for consumers and not decision-makers.

[Biopolitical]

Nice post and nice comments.

You have already mentioned two problems with BCA guiding policy decisions. First, government bureaucrats are likely to do BCA ineptly. Second, decision makers incorporate other parameters, including their own preferences, as they see fit.

My alternative is to have freedom from would-be policy makers, and to have markets and property rights. But don't trust me -- I am, alas, just an ecologist.

[Dave Iverson]

Robert,

You say or ask, "I haven't read Policy Paradox and A Primer on Decision Making] but I assume they advocate multi-criteria analysis (MCA)?"

No, they do not. James March and Herbert Simon might be called the fathers of "bounded rationality," which is a way of looking at the world in non-closed form to help people make decisions and learn to better their decision-making. March's "primer" is in that vein.

Deborah Stone's book too comes at the problem from this perspective. That why she framed the book and the title around the word "paradox."

Read one or another and see what you think. Read them both and let me know what you think of them. You might be pleasantly surprised at how much of the world operates outside the frame of what Stone criticizes as "the rationality project."

And if you like either of these, then you can step into Robert Jervis' Systems Effects: Complexity in Political and Social Life that also frames things up this way, dealing with national security and other world-scale contexts as stage setting for helping us to better understand the dynamics (complexity and political wickedness) of human systems.

Here is a snippet from Stone's preface:

"The broad goals and principles at the heart of political conflict — things such as equity, efficiency, liberty, and security — can never be reduced to simple deterministic criteria, and therefore they can't tell us how we should best decide policy questions. ... For all the trouble caused by vague goals, imprecise problem definitions, and unruly policy instruments, we would be fools to trade them in for a calculator."

[John Whitehead]

Dave,

Is there anything that I could read in an evening?

John

[Dano]

IME, folk that don't want BCA as a metric usually complain because BCAs assign measurable valuation to ecosystem services, & that makes it more difficult to exploit ecosystems and/or to dump their externalities on them (viz. "polluter pays" laws):

First, government bureaucrats are likely to do BCA ineptly. Second, decision makers incorporate other parameters, including their own preferences, as they see fit.

Decision-makers don't do BCAs. They have folk do it for them. This is a non-starter.

Second, decision-makers incorporate BCA in addition to other parameters. The marketplace of power relations determines those parameters that are in play.

My alternative is to have freedom from would-be policy makers, and to have markets and property rights. But don't trust me -- I am, alas, just an ecologist.

You would just replace current policy-makers with new policy-makers, as Homo sapiens throughout geologic history has organized itself into social groups with leaders.

[Dave Iverson]

John and Others:

Interesting challenge: "Is there anything that I could read [re: March and Stone, etc] in an evening?."

Trying to understand the literature of managerial and political decision making in an evening is about as much an impossibility as trying to understand the history of economic thought in an evening. Even to begin to understand is tough.

Neither Stone nor March, to my knowledge, has packaged their materials so as to be rendered into "one minute" or "one evening" packages.

I did run into this little powerpoint, however, that is titled "A Brief History of Decision Making" [PDF] that may be of some help.

It hightlights some of James March's contributions, along with those from some game-theory and decision theory economists. It is drawn from a Leigh Buchanan and Andre O'Connell article from Harvard Business Review, Jan. 2006.

[allen claxton]

Dave, those books, judging from their amazon pages, look more like descriptive accounts of decision-making processes, rather something that provides tools to use in those processes, which is what BCA is. Am I being fooled by their covers?

[Dave Iverson]

Allen,

They are not "cookbooks" if that is what you desire. But they do give you enough insight to begin to better understand decision making/policy making process and context, and to begin the decision-betterment process.

One punctuation point from this school of thought is that there are no easy answers, no silver bullets, no magic potions. Life if fraught with uncertainty and risk, it is filled with paradox. Life is best lived with some understanding of the dynamics that suround us in the social, biological, and natural systems that enfold us and unfold before us and the future emerges.

[John Whitehead]

Dave,

Can't people use benefit cost analysis and understand all of those things?

John W.

[allen claxton]

They are not "cookbooks" if that is what you desire.

No, but when I see someone ask for alternatives to BCA, I think of stuff more like life cycle analysis (not so much for policy analysis, admittedly) or ecological footprinting. Not so much as replacements of BCA, but complements to it. But that kind of tool, rather than a general overview of how decisions get made.