Secondly, a longer treatment can be found here -- an entire downloadable book titled "What Environmentalists Need to Know About Economics".

As The Economist magazine remarks (quoted on the linked page) "Academic disciplines are often separated by gulfs of mutual incomprehension, but the deepest and widest may be the one that separates most economists from most environmentalists...What underlies this is not so much disagreements about facts as disagreement about how to think."

I do hope this blog serves to lessen that disagreement to some measure. I'm also pleased to see other economists (like Jason Scorse, the author of the book linked to above) writing explicitly for non-specialist audiences.

1. How to think about how economists think

First I thought about analogies that non-economists might base their impressions of what economists do on. One analogy is the “machine” analogy—that non-economists think of “an economy” as a machine with moving parts that economists learn to understand and tweak. (This positions economists as engineers, or mechanics of some sort.) An alternative analogy is where the economy is like an “organic machine” (like a human body), with economists being like “doctors” who diagnose and prescribe. These metaphors tend to encourage a view of economists as those who “fix” a machine, whether mechanical or organic, when it breaks down. (I ought to add, I first prepared these notes before the onset of the Global Financial Crisis, in which the idea of economic “break-down” rose high in prominence.)

What I said to the assembled ecologists was that these analogies are not good representations of what economists do and how they think. It’s better to regard economists as looking at a system that has independent self-seeking agents within it, responding to changes in circumstances (including policy settings and actions of other agents). Ecologists should not find this conceptually unfamiliar, given they typically examine systems that respond to outside pressure, with individual agents (species) responding in ways that can be understood using Darwinian principles.

One of the things economists have in their heads is the idea that a given configuration of a system produces a certain amount of “value”. Economists are interested in how different configurations of the system will change the value being generated. Note that while human economic activity generates value (captured in e.g. GDP), value can also be generated by nature in various ways, and that value-generating potential depleted or destroyed by human activity. This is the “space” in which ecologists and economists can fruitfully collaborate.

A potentially important distinction in how economists and ecologists instinctively look at the world involves the relative importance of theory versus data. Economics—even applied economics—tends to start off deductively, referring to theoretical principles before moving to data analysis. Applied scientists often proceed fairly inductively, perhaps with reference to some broad (e.g. evolutionary) first principles and some micro laws. This is fine, in principle, as long as both groups understand this about the other.

2. What economists can (and do) do

One thing economists do is to “measure value”—in simpler terms, “look at the dollars” involved with some proposed aspect of scientific research or policy intervention. Since in the environmental/ecological space, we are talking about being outside the market sector, we are typically talking about “virtual dollars” here. That is, “measuring value” in this context will typically involve what is referred to as non-market valuation.

(Example: I may be able to assess the financial upside to developing a wilderness area and building homes or factories there. Whatever practical forecasting difficulties I face in answering this question, the question itself is simple in principle to answer. The harder question is to assess what the wilderness area is worth being left as it is? This value will normally not be captured in market transactions, but that doesn’t mean the value doesn’t exist.)

Another job economists do is to “measure cost”—this is one part of measuring value in net terms, but in some cases the upside may be too hard to measure. However, economists can look at alternative proposals to achieve a given end and assess comparative costs of those proposals, taking the benefits as given.

Another is “policy assessment” (or “policy design”)—looking at whether a given policy action is likely to have the intended effect. Recalling that in a system comprised of self-seeking agents, they will respond to changes in e.g,. policy settings, potentially in ways not intended by the policy-makers. Hence, policy proposals need to be assessed in light of the Law of Unintended Consequences. Related to this is policy design, in which policy interventions are designed from the ground up to be (for example) “incentive-compatible” and “cost-effective”.

3. Metrics matter

So … we can talk about “systems” that create “value” in different ways, in different configurations, subject to various shocks and pressures. These systems have agents within them who behave in (somewhat) predictable ways in response to those shocks and pressures. (Economists will typically talk about changes in incentives, but that can be regarded for our purposes as simple semantics.)

How members of different disciplines measure things and then predict outcomes can heavily influence our analysis, and hence the recommendations we might make in policy documents and discussions.

Scientists tend to work with physical variables expressed in physical units—metrics—and do their analysis and draw their conclusions based on these. Economists tend to work with a more broadly defined and “dollarized” value kind of metric, even when talking about physical impacts. (I could talk here about economic efficiency, but maybe another time.) Working as economists do tends to lead to different conclusions and prescriptions than if one only worked in physical metrics.

Example 1: Maximum sustainable yield.

If you were asked to give a “sustainable harvest” recommendation over, say, a marine ecosystem (i.e. a fishery), what might you do? A marine scientist (and we’re talking about a hypothetical person here, nobody I know) may compute different sustainable harvests associated with different stock sizes. That is, compute how much a stock of given size grows by in a time period, and let that net growth be the amount that can be harvested. (It keeps the stock constant, hence the harvest is “sustainable”.)

The next step for our hypothetical scientist, after doing this for stocks of various sizes, might be to identify the stock size that results in the maximum sustainable yield—for a standard stock growth function, there will be a stock size that results in maximum stock growth, enabling the largest sustainable harvest. If harvest size is the metric, that’s the right recommendation.

But it’s unlikely to be the recommendation that results in the greatest value from the fishery. The cost of resources devoted to catching fish—“fishing effort”—will vary for different catch sizes, and if you don’t factor that in, you will make a recommendation that fails to maximise the net benefits—value—from the fishery. Focussing only on the physical metric means you maximise the gross value, by maximising the size of the catch.

Example 2: Energy efficiency.

What if a scientist or engineer came up with a way to increase energy efficiency by 5-10%? Is that cause for celebration?

Maybe. You can bet that if it was a University of Sydney researcher, this news would be touted far and wide, and hailed as a great success of the new Institute for Sustainable Solutions by the VC, Provost, and whoever else could make a noise about it.

Meanwhile, the economists would be looking at the finding and saying “Hold on a minute…”

The problem (such as it’s a problem) is that increasing energy efficiency (a physical metric) decreases the cost of using energy (a value metric). This in turn is likely to increase the use of energy. Remember that we’re talking about a system in which agents respond. They’ll respond to things like changes in costs.

Therefore, the impact of increased energy efficiency on actual energy use is an empirical issue. A 5% across the board energy efficiency increase will almost definitely result in a less-than-5% drop in energy use (the gap is called “rebound” in the literature), because of the drop in energy costs. This is likely to occur mostly in the production sector as businesses substitute more (cheaper) energy for other inputs. If the effect is large enough, the net effect is that energy use could in fact increase, a very unexpected physical outcome that could not be predicted by focussing only on physical metrics (this is called “backfire” in the literature, for obvious reasons).

Backfire in energy use is a very strong result to expect from an increase in energy efficiency, and would only result from very particular assumptions about production structures (what economists refer to as input demand elasticities and so on), but the rebound effect is almost certain to exist. Which is to say, a 5% increase in energy efficiency will not translate into a 5% reduction in energy use.

So, while we need to focus on energy efficiency as part of a set of sustainability strategies, we have to remember that making our use of energy more efficient means making energy more attractive to use, which has follow-on impacts.

4. Case studies

Four of us discussed case studies of environmental/ecologic economics research that has used (and could use more) interaction with natural scientists.

I discussed work I’ve been involved with on the economics of ecological resilience—how to think about (and quantify) resilience as a component of overall social wealth, and assess changes over time in resilience and the impacts on net wealth. I also mentioned work done in the area of Market Based Instruments, where market instruments are used to achieve environmental outcomes—in one famous case (Victoria’s BushTender) scientists constructed an “environmental net benefits index”, and economists worked out how to “maximise bang for the buck” from the environmental auctions.

Tiho Ancev spoke on determining economic benefits from environmental water flows based on ecological benefits provided (which appear to be vary substantially across time), and also about research on water quality management (algal blooms) for large dams that cater water to urban water supply.

Greg Hertzler discussed work in the area of bio-economic modelling. He gave an example involving the management of elephants as a potentially endangered species, and made clear how the results were sensitive to assumptions in the modelling process (in particular, looking at raw biomass, versus an age-structured population).

Bob Cairns of McGill University spoke about the economics of the ecological footprint, a measure that has been widely critiqued by economists for being ad hoc and focussed only on physical metrics. Bob is seeking to work from first principles to understand in what setting an ecological footprint measure might yield meaningful information.

And, as an economist, it’s not at all unnatural to me to see exactly what I’ve described above – economists at the forefront of dealing with what I see as fundamentally economic problems. One of the journalists (Glenn Milne, I think) on a recent Insiders program on the ABC said, almost as an aside, that climate change was an economic problem. And it is. If climate change was happening but had no significant consequences, then it wouldn’t be a problem, merely a scientific curiosity. The fact that it’s a problem means it has consequences that can be represented in terms of costs and benefits (as can policy actions designed to mitigate and/or adapt to climate change). Being “a problem” means that it’s an economic problem, more or less by definition.

Economists have thought about a variety of these kinds of problems for some decades now, in various forms. (What are now widely discussed as “cap-and-trade” schemes, such as for emissions trading, have been discussed by environmental and resource economists for over three decades now.) What’s been missing, or at least somewhat marginalised, for a long time, has been high quality collaborative multi-disciplinary work where scientists and economists came together to tackle key problems. Such work is not unknown, by any means, but it’s still difficult to get people working outside (or across) their disciplinary boundaries, and to put their own biases and disciplinary prejudices on hold.

And yet, now more than ever, it’s crucial that scientists and economists come together. Economists can only get so far looking at important environmental/ecological/natural issues without getting information on key parameters and interactions from the experts in the area – the scientists.

In one sense, what I’m saying is trite. Neither Stern nor Garnaut could do what they’re setting out to do without building on the scientific work of the IPCC and others. I talked about other examples in December with some colleagues, but the simple moral is, leading-edge work by economists on natural resource and environmental issues relies on scientific underpinnings. When the stakes are high (and when the government is commissioning something), good economic analysis drawing on science will emerge. The problem is, in my view, that there’s still far too little integrated, collaborative, multi-disciplinary research occurring between economists and natural scientists.

So how to make this happen? Economists need to be willing to better explain what it is they do, why they do it, and where the big unresolved issues still are. (One motivation for starting this blog, of course.) Scientists probably need to accept that while “good science” is a necessary condition for “good policy”, that is not the same as being a sufficient condition. Putting it as simply as I can, science will lay out our available options for us. Economics helps us make choices from those options. (There are other things economists can contribute, as well, but that is a big one.) As long as the world is a complicated place, both the scientific analysis and the economic analysis will be contestable, but that’s life as a researcher. My point here is to simply identify what I think is a reasonably appropriate scientific division of labour, as a prelude to what will be a bit of a campaign to enhance the scope for productive research collaboration between scientists and economists, because in natural resource and environmental policy, that’s increasingly where the action will be.

I’ll talk a little more about the session at the ESA in a separate post, but for now I’ll note that my Faculty colleague and fellow blogger Dr Willem Vervoort has commented on some discussions I was having with economist colleagues about how we might approach the session. Willem has been an “early adopter” of collaboration with economists within the Faculty, particularly in the form of a series of co-supervisory arrangements for graduate students. I’ve written too much already to address any of his comments in detail in this post, but I might come back and do that in a separate post. Suffice to say that I agree with a fair bit of what he says, but I’m going to reflect on it a little bit more first before I decide whether I want to clarify anything.

The debates around the scientific issues—and the extent of the sheer uncertainty surrounding the potential impacts—would bedevil anyone trying to do a Stern-type exercise. But here we won’t focus on that. Instead, the goal here will be to come to grips with Stern’s appraisal as an economist of the impacts and implications of climate change—and what other economists found to disagree about in Stern’s assessment.

Since an economist studying aspects of climate change would find plenty of particular topics and individual problems open to study, we should clarify what is meant by “the economics of climate change” as looked at by Stern. His was a big-picture analysis of the long-term global costs of climate change, and the potential for reducing those future costs by taking action now.

What caught people’s attention—and raised eyebrows—was that Stern found the costs to be considerably higher than most previous estimates, and based on this, he made a strong case for action now. There were, in short, two reasons for this finding. First were the scientific assumptions that went into the analysis: what would be the physical changes expected under warming, and what would their economic impacts be expected to be? Second, taking the first assumptions as given, how does one convert a whole stream of economic impacts (net costs of warming, and net benefits of action) over a long period of time and convert them into a single number—a present value—for purposes of comparison?

In short, oversimplifying somewhat, we can think of the first part as “getting the science right”, and we can think of the second as “getting the present value right (in practice, choosing the discount rate)”. Stern’s assessment of the science was dramatic—the impacts of warming would be big, and they would be bad—and was one reason for his result that the “costs of doing nothing” were significant. However, his discount rate choice was also a big influence in Stern’s final quantitative conclusions. It’s the second debate that occupied the attention of many economists, and where we shall focus our attention here.

This means I am not going to focus on whether his estimates of the costs measured over time are accurate—whether he underestimated the costs associated with Arctic ice-sheet melting, or miscalculated the risks of sea-level rise, or of changes in agricultural productivity. Instead I’ll focus on the discounting issue, because that’s where the economics profession got really fired up. And to understand the debate, we need to comprehend just one equation.

This equation, attributed to Frank Ramsey (Wiki link here), summarises the rationale for discounting in a parsimonious way, combining a few key variables. But before we unpack the equation, we need to talk about what discounting entails.

Imagine the following scenarios. These are totally made up, of course, but they are useful thought experiments to use when thinking about the issues raised by Stern.

In one scenario, you take $100 and put it in a hole in the ground. You come back the next day, dig it up, and find yourself with $1000!

In another scenario, you take $100 and put it in a hole in the ground, and you come back in 50 years (you can’t do this earlier!), to find yourself with $1000 when you dig it up.

In a third scenario, you bury $100, and the next day, somebody entirely separate to you comes along and digs it up to get $1000.

What these all have in common is that they are “investment scenarios”, where money set aside (diverted from being spent immediately on consumption) comes back as a larger amount.

What’s different about investment scenarios one and two? Perhaps nothing, right? You bury $100, which somehow magically grows, and you get back $1000. The only difference is that in one case the payoff is the next day, compared to the other one taking fifty years. Does this matter to you? It’s hard to imagine that it wouldn’t. If I knew that I could earn around 4.8% (real, meaning inflation-adjusted) per annum on an investment, then $100 invested now would yield me a bit more than $1000 (inflation-adjusted) in fifty years time. That’s OK, but it’s hardly amazing. I don’t think it’s obvious that I should divert every dollar I could into an investment that paid less than 5% p.a. However, the overnight investment is amazing! I don’t even know what that rate of return that would be per annum, because Excel spat the dummy when I tried to calculate it (it’s 900% per day). But if I could get $1000 overnight for every $100 I set aside, you can believe I’d be living on bread and water while I crammed every spare dollar I could into the Magical Money Expander in order to get it back tenfold the next day.

The point of these thought experiments is that money that arrives later is worth less to us, even when we adjust for inflation. The $1000 payoff in fifty years time is simply not as attractive as the $1000 payoff tomorrow. Not even close.

This provides a prima-facie rationale for discounting future dollars, such that at around a 4.75% rate of discount, $1000 received in fifty years’ time is the equivalent of approximately $100 today. In the lingo, $1000 in fifty years has a present value of $100 today (when discounted at 4.75%). (Note—and this is important where Stern is concerned—that the rate I use to discount matters a lot. The present value of $1000 in fifty years would have a present value of $290 if I discounted at 2.5% instead of 4.7%. Which means, working forward in time, I need almost three times the initial investment to receive a given payoff in fifty years.)

What about scenarios one and three? Now the difference between the payoffs is not when they occur, but to whom. Now, the Magical Money Expander takes your $100 and creates $1000 from it, but somebody else—some random person that you don’t know—gets the proceeds, not you. Under what circumstances would you think it is reasonable for you to give up $100 in order that someone else receives $1000?

Now think about it from the point of view of the Hypothetical Benevolent Social Planner (HBSP), an imaginary person whose job it is to make (potentially major) changes in society for society’s overall benefit. The HBSP can take $100, stick it in the Magical Money Expander, and create $1000 of benefit for someone else other than you. If the recipients were poorer than you, would you think it a reasonable sacrifice to make—you lose $100 for some poorer people to get $1000? (The HBSP might think so!)

What if the recipients were richer? The HBSP would need to trade off the fact that $100 has been turned into $1000 (a good thing), while the transfer involved was from poorer to richer (let’s call this a bad thing). In what circumstances might a HBSP think the sacrifice was still worth it? What if the Magical Money Expander only turned $100 into $150, instead of $1000? Would the increase in overall wealth compensate for the fact that the rich have received a transfer from the poor?

All major public investment issues—involving benefit-cost analysis—involve transfers of resources across time, and among different people. (I’ve represented these here in simplified fashion using thought experiments involving an imaginary Magical Money Expander.) It is benefit-cost analysis that helps make sense of these transfers; to decide whether the effects of the project in question (the “Magical Money Expander”) create sufficient benefits, once the effects of transfers through time and between people are properly taken into account.

In this situation, climate change mitigation action plays the role of the Magical Money Expander (to the extent that undertaking global mitigation actions will require sacrifices now that will result in reductions of large costs of warming that we would otherwise incur in the future), and Stern plays the role of the HBSP. The transfers that Stern considers occur over large spans of time, large enough that the main beneficiaries of our costly actions now will be people not yet born. And, by and large, those people will also be the beneficiaries of the economic growth that occurs between now and then, so they will be, on the whole, richer than us. Whatever sacrifices we make (for a problem we have contributed to, but that really kicked off well before we came along) will mainly benefit richer people, well into the future. How should Stern apply discounting in this situation?

Let’s go back to the Ramsey equation.

The left hand side, p, is the rate of discount to be applied to future net benefits of mitigation actions taken now. It is the sum of two terms on the right hand side which are governed by two key parameters. The first, the delta, is the “pure rate of time preference”, dealing with our scenario two above—the fact that the benefits arrive later in time relative to the costs of the sacrifice that generated them. The second term is the multiple of two measures. The eta symbol is the “inequality aversion parameter” that deals with scenario three above—the fact that most of the costs of mitigation will be incurred by people poorer than those to whom the benefits will accrue. The term g measures economic growth, more or less.

(You could argue that the future impacts of climate change, if unchecked, will hit poorer countries harder, and so we in the West are not transferring to our richer descendants by taking mitigating actions; rather, we are, by mitigating, making progressive transfers from richer to poorer countries. But then, should we be taking $100 from citizens of richer countries today and turning it into a $1000 benefit for citizens of poorer countries in 50 years’ time—or should we just give that $100 to the citizens of poorer countries now and let them do what they want with it?)

Re-capping: Stern found high future damages resulting from climate change, based on comparatively heavy emphasis on future downside risks. But the significant point that fired up economists was that he found high present values of damage, such that there was a clear case for undertaking actions now. That aspect was driven in large part by Stern’s choice of discount rate p, which involved in turn making choices for delta and eta. The choices Stern made were low, meaning that p was on the low side, which means future impacts were not as heavily discounted as they might have been. The long time frames involved meant that the choice of a low discount rate had a major influence on his conclusions. This meant that the choice of a low discount rate was a prime area for controversy.

Dasgupta, P. 2007, ‘Commentary: The Stern Review’s economics of climate change’, National Institute Economic Review, vol. 199, pp. 4–7.

Jensen, P. and Webster, E. 2007, ‘Is Stern Correct? Does Climate Change Require Policy Intervention?’, The Australian Economic Review, vol. 40, no. 4, pp. 421–31

Nordhaus, W. D. 2007, ‘A review of the Stern Review on the Economics of Climate Change’, Journal of Economic Literature, vol. 45, pp. 686–702.

Stern, N. 2006, Stern Review: The Economics of Climate Change, viewed January 2007, http://www.hm-treasury.gov.uk/independent_reviews/stern_review_economics_climate_change/sternreview_index.cfm.

Sterner, T. and Persson, M. 2007, ‘An even Sterner Review: Introducing relative prices into the discounting debate’, Resources for the Future Discussion Paper RFF DP 07-37, Washington, DC.

Weitzman, M. L. 2007, ‘A review of the Stern Review on the Economics of Climate Change’, Journal of Economic Literature, vol. 45, pp. 703–24.