In september 2000 (the 23rd to be precise), a paper published in Le Monde presented the externalities due to CO2 emissions for road transportation of goods (an externality is a cost that is not directly paid by the user or the client. For example road accidents, that have a cost which is only very partially paid by the drivers, through insurance premiums, the rest being paid by the various welfare systems or the federal budget, constitute an externality of car use).
Does it have any sense to give a figure for the externalities resulting from greenhouse gases emissions, that induce climate change ?
Indeed, there are two things that should be distinguished albeit they are often mixed up: the cost of avoiding a damage and the cost of reparing a damage. In the above mentionned article, the journalist agregated the cost of infrastructure reparation (being much heavier than cars, trucks damage infrastructures much more than cars) with a “greenhouse effect cost”, without any further details.
A profane reader will probably consider that the amount mentionned – that was modest, as a matter of fact – should be paid by the car and truck drivers to feed some kind of “savings account” that would be used, when the day comes, to repair the damages caused by climate change, just like the “infrastructure reparation” externalities correspond to the amount required to repair roads when too many vehicles have circulated on it. This approach implies that damages induced by climate change can be given a monetary or economic counterpart, or that a damage cost can be calculated for climate change.
Well, estimating an objective damage cost for climate change is not the least possible, because the mathematical expectancy of such a ccost is infinite.
Maybe, before going any further, it might be appropriate to recall what is a mathematical expectancy. It corresponds to the intuitive notion of what happens “on average” when one repeats the same action a large number of times. When, on average, I need 15 minutes to go from point A to point B, I can expect to spend about that time the next time I make the trip, and therefore I will consider that the “mathematical expectancy” of my travel time is 15 minutes: it is the most probable time that I will spend in the future given what I spent in the past.
The mathematical expectancy is therefore the average of all the possible values of a future event (and it is always the average value that we expect to find). By definition, it is worth
{\mathbb{E}[X] = \sum_{i=1}^infty x_i, p_i,}
Sum, for all possible events, of
[value of the possible event × probability of the possible event]
One will immediately see from this formula that if one of the terms has an infinite value, with a probability of occurrence over zero, whatever small the latter is, the whole mathematical expectancy is infinite.
All risk management is based on mathematical expectancies. For example, insurance premiums correspond to the mathematical expectancy of the possible damages insured (plus the expenses and the profit of the insurance company, of course !). What we intuitively call a “risky behaviour” is nothing else than a behaviour for which the mathematical expectancy of the possible damage cost is high (or at least over a certain threshold). It means that there exists, in the list of the possible consequences of such a behaviour, at least one that has a high cost of damage, along with a significant probability of occurring. Reckless driving is considered risky because one possible event (its probability is over zero) is that the driver dies (the associated cost is high), and that pulls up the whole mathematical expectancy of damages of driving like mad.
If we come back to climate change, in order to justify with economic arguments that it is “profitable” (or economically sound) or “not profitable” to act against it, it is necessary to get a figure for the economic cost of a number of possible consequences, along with a probability of occurring of the said consequences, to get a mathematical expectancy of the damage cost. It is the comparison of the “damage cost” of these consequences with the the cost of prevention that will allow to choose what is best. Furthermore, action is for now, when possible damages are for later, so that it is necessary to “actualize” the cost, that is choose an abatement of the value because it concerns the future. But to perform such a calculation, we have a major problem ! Indeed, what “price” can we attribute, in 2004, to the following events:
- the death of 30% of humanity in 2089 because of the extent of tropical diseases ?
- 10 meters extra for the ocean level in 2350 ?
- The desertification of half the planet in 2080 ?
- A turning of the carbon sinks into sources in 2055, and the death of 99% of men in 2200 ?
Of course, no economic model is able to give a financial value to such events, not even an upper limit: it all depends on hypotheses ! We might say that the cost of such events, if they occurred (most of all the last one !), is infinite, since no-one can give them an upper limit. Nobody either can state that these events are totally impossible, therefore their probability of occurrence is not zero.
As a consequence of the fact that one of the future possibilities has an infinite reparation cost, the mathematical expectancy of the reparation cost is infinite also, whatever the probability of that event (as soon as it is only possible), and even whatever the actualization rate associated to this reparation cost (since it is a future expense). Hence it is not possible to say, today, that it is not “economically sound” to act against climate change.
For those that are not afraid by corporate accounting: let’s laugh a little
Any corporate accounting must be prudent. Suppose that the legislator (or in France even the judge, since the principle of prudence is already part of the law), acting wisely, obliges any company that uses fossil fuels – that is any company, in fact, since having heated offices is enough to emit greenhouse gases – to provide, in proportion to its cumulated greenhouse gases emissions, for a future “restoring of the climate system” (such a rule would have a true meaning, since this company contributes to degrade the climate system, in proportion to its cumulated greenhouse gases emissions).
This is equivalent, in a way, to attribute to this company a share of the reparation cost of the future damages, with no actualization. As there is no objective limit to the cost of the future damages, as stated above, there is no objective limit to the amount of the provision, that an accountant could set at a level that would lead to a loss for any company. Still, would it be such an absurd rule, theoretically speaking ?
But for this problem we can still define a “cost of avoiding”, or of dissuasion, just like for tobacco : the name of the game is then not to put money in a savings account to pay for future damages, but to dissuade an economically rational agent to emit a lot of greenhouse gases.
The carbon tax is based on such a prevention principle. However, if a tax is set up someday, its value will probably be more the result of harsh negotiations than of a precise quantification of the expected effect, as it is often the case.
A little glossary : what exactly is economy?
One of the reasons why some people still go on saying that acting against climate change would be “bad for the economy”, in spite of all what is exposed above, is simply that the word “economy” can be given a lot of different meanings. Indeed, depending on the context and who’s talking, “the economy” can designate:
- the turnover of the company that employs who’s talking. ” It’s bad for the economy” then means: “it’s bad for the turnover of my company”, but one company’s turnover is not the whole economy ! And what is bad news for some is sometimes good news for others…
- the net profit – or the share value – of the company that employs who’s talking. ” It’s bad for the economy” then means : “the value of the share might be pushed down” (or even worse: the value of my stock-options will decrease” !) , which is not the same thing already,
- The number of staff in the company that employs who’s talking. ” It’s bad for the economy” then means: “The number of jobs in my company might decrease”, or “the number of qualified jobs might decrease”, or “I will increase more slowly the number of jobs”, or even something else… (but the share value might not be affected, nor the turnover, nor the number of jobs of some other companies….), any of the above notions for a whole activity sector. For example, when a head of an union trade for steel manufacturers says “It’s bad for the economy”, he certainly means it for the turnover of steel mills, of the number of steel workers, but are we sure that he ALSO means it for potato farmers ?
- The productive activity of a whole country, that is generally characterized by the Gross National Product (GNP) One should know, still, that GNP is counting flows, not assets: it represents the cumulated added value of all merchant and non merchant activities located on a given territory. With a couple of adjustments (stock movements, net balance of the exchanges with foreign countries), the GNP corresponds to what we spend. It therefore measures human activity, but in no case a patrimony ; stating that we are richer when the GNP increases is thus a big nonsense (just regarding the meaning of words, the rest being a vast debate): do I get richer when I buy more sweets ?
In addition, one should note that what is good for the GNP is not necessarily good news as a whole: increases in weapons sales, in de-pollution services, in disease cure, etc, all increase the GNP. A country with lots of enemies at its door (good for the military business), polluted water (good for the de-pollution services), obese people (good for medical activity, food manufacturers, car manufacturers because they can’t walk, etc), noisy suburbs (good for manufacturers of soundproof material), stressed men (very good for the pharmaceutical industry), etc, has a flourishing GNP, but does it correspond to the idea we have of a flourishing quality of life… ?
On an even more global point of view, if we used an “assets and liabilities” approach for states, which is generally not the case in most national budget accounting (and which is the proper one to conclude that a company or an individual is getting richer: its assets must then increase faster than its liabilities), aggregating “artificial” assets (houses, roads, software, plants, etc) and “natural” assets (oilfields, forests, arable land, rivers, the climate system, etc), as we generally degrade the latter in order to increase the first, it is not possible to formerly conclude that we are currently increasing our global wealth in the sake of economic arguments.
- at last, “the economy” might designate the average level of material consumption of the population, which, ironically, will often not be expressed in economic units, but in consumption units: number of cars sold, of square meters of houses built, etc. And here it’s true that, in a world that would have decided to make climate change mitigation a priority, everything that is expressed in physical units will have a hard time growing forever (whether we like it or not !). But it would be more exact, then, to say that acting against climate change is bad for the growth of our material consumption (which is definitely true, but not necessarily bad news: this kind of growth goes along with the growth of obese people, stress, traffic jams, noise, etc). But then we are talking of a particular aspect of “the economy”, which is the material consumption expressed in physical units, not of economy as a whole!
“The economy” can also designate the balance of the foreign trade, the balance of financial exchanges, the GNP growth rate, or the unemployment rate…. Well, when someone affirms that acting against climate change is bad for the economy, even if the someone is a Nobel Prize, what economy exactely is (s)he talking about, and what sound numerical hypotheses can comfort this statement ?