top of page

Chapter 7

Just dig it!

La Cruz de Huanacaxtle, October 23rd, 2017


What do Bernie Sanders and Rex Tillerson have in common? They are both members of the Pigou Club.1 Members of the Pigou Club have once publicly argued in favor of a Pigouvian tax, named after Arthur C. Pigou. It is a tax to correct for externalities; for example, a tax on carbon emissions equal to the external damage. In addition to its environmental benefits, the club manifesto argues that these Pigouvian-taxes would ease road congestion, regulatory burden, and the possibility to keep the total tax burden equal by lowering other taxes.2 So, why don’t we raise more Pigouvian taxes? While popular among economists, taxes on gasoline, flights, or meat are unpopular with voters.3 

The idea of the Pigouvian tax may have gotten to Pigou while he was pondering over the puzzle of economics as worded by Jevons (1879, p. 289):4 “Given, a certain population, with various needs and powers of production, in possession of certain lands and other sources of material: required, the mode of employing their labour which will maximise the utility of the produce.” Put differently, what distribution of efforts across people, lands and materials maximize social welfare, and as the puzzle is formulated rather generally, one could easily replace labour by dollars. According to Pigou (1920, 2.2§5), to maximize social welfare one should account for externalities, or in his words: “costs are thrown upon people not directly concerned, through, say, uncompensated damage done to surrounding woods by sparks from railway engines. All such effects must be included—some of them will be positive, others negative elements—in reckoning up the social net product of the marginal increment of any volume of resources turned into any use or place.5

Perhaps the former paragraph was a bit abstract for your taste, both Jevons and Pigou were masters at math as well as using almost mathematical formulations.6 So, let me return to some examples: I first introduced the concept of externalities as the difference between the price and the cost of a flight. I described that by not pricing the uncompensated environmental damage, or social cost of carbon (SCC), people fly too often from a social perspective. An experiment by Gneezy et al. (2004) illustrates how externalities influence consumption behavior, rather literally: The experiment compared diners in a ‘split the bill’ treatment with a control group, where diners paid for their own meal.7 Sharing the costs of one’s meal between a table of six, leads diners to eat for about a third more expensive. As the costs of carbon are shared with the planet, including other species and future generations, the overconsumption of carbon is probably worse than by a third. Isn’t it odd that the status quo is such we split the bill on each other’s carbon consumption?

Pigou proposed a tax equal to the uncompensated damage as a solution to this externality problem, so people would ‘internalize their externalities’. In other words, the tax leads people to take the otherwise uncompensated damage into account. This solution is now widely known as the Pigouvian tax.8 When governments tax the emissions from flights at their SCC, polluters will pay for their environmental damage or are discouraged from flying. Either way, people start to arrive at the social optimum by themselves, by internalizing the environmental damage in their decisions. The tax improves welfare by dissuading some people from flying, those for whom the private benefits outweigh the price of a ticket, but not the cost. Others will still choose to fly, in this case their benefits outweigh the price plus SCC, but at least the tax will help governments raise funds.9 

Our relationship has had a bit of rocky start. Since I arrived in La Cruz, she did not pay much attention to me, but last week at the beach, I think she lost one of my flipflops out of spite. Probably she was either jealous of the attention I get from Alef or perhaps she sensed that I’ve never been much of a fan of dogs. So far, my favorite dog had been the tv-character wearing a dog-suit Wilfred rather than an actual dog, but I admit that Luna is starting to grow on me. According to Alef, the flipflop incident meant that I was growing on her as well. I guess dogs just have funny ways of showing their affection. Sometimes their way of showing affection is more painful than annoying; Luna is still ‘gently’ biting and jumping us when she gets too excited, like every time we get to the beach.

Alef is not sure about the race nor about the age, because she adopted Luna about a year ago, but Luna seems a mix between a Labrador and a Collie and about three years old. She is all white apart from some black spots on her ears. Alef adopted Luna after a client brought her into her restaurant. The client had told Alef that she needed to return to Canada and that she could not bring Luna to the pound, because Luna was already on the pound’s ‘third-strike-and-you’re-out’-list. Luna has been somewhat of delinquent, having been returned to the pound twice after adoptions, the second time for killing a chicken, allegedly. Despite being aware that her lifestyle is unsuitable to have a dog, Alef took her in with the intention to find a new home for Luna. Alef soon discovered why Luna had been returned so often: Luna is deaf, which makes training her more challenging.10

Luna has had a troubled past and the scars to show for it. Alef told me that Luna had nightmares in the first months after her adoption. The nightmares could be about losing most of her tail or about the time she was living in the streets. Fortunately, she doesn’t wake up barking lately, but perhaps her troubled past explains why she gets so anxious whenever we leave her home alone. We hear her bark until we leave the street, and usually, Luna makes us pay for not taking her with us. Sometimes she takes down the mosquito screens from the windows, other times she empties the trashcan in the living room. The Wilfred series had portrayed dogs as digging out of anxiety, and since we don’t have a bedroom door, Luna did just that with our foam matrass. We had to play Tetris with the leftovers from our matrass twice over the past week.


Regarding the social cost of carbon (SCC) estimate of $40 that I mentioned in my last letter, let me start by explaining how the U.S. government’s Interagency Working Group arrived at their estimate: “The most common approach to estimating the SCC uses an Integrated Assessment Model (IAM) or related model to simulate time paths for the atmospheric COconcentration (based on an assumed path of CO2 emissions), the impact of the rising atmospheric CO2 concentration on temperature (and perhaps other measures of climate change), and the reductions in GDP and consumption that will result from rising temperatures. The idea is to perturb the assumed time path for CO2 emissions by increasing current emissions by one ton, and then calculate a new (and slightly lower) path for consumption. The SCC is then the present value of the reductions in consumption over time resulting from that additional ton of current emissions (based on some discount rate). Note that the SCC calculated this way represents the marginal external cost of emitting an extra ton of CO2.11

I have to disappoint you on how consistent these SCC estimates are: “estimates of the social cost of carbon (SCC) range from around $10 per metric ton to well over $200 per metric ton, and there has been little or no movement toward a consensus number.12 Pindyck (2017, p. 348-9) goes on to explain why it has been so difficult to move towards a consensus: “Even with no uncertainty, the time horizon makes the present value of future benefits from current abatement extremely sensitive to the choice of discount rate, and there is considerable disagreement over what the ‘correct’ discount rate should be. And then there are the very large uncertainties, some of which we cannot even characterize. The more important uncertainties pertain to the extent of warming under current and expected future greenhouse gas emissions, as well as the economic impact of any climate change that might occur.13 These ‘more important uncertainties’ are also referred to as climate sensitivity and the damage function, respectively. Uncertainty about climate sensitivity (or the magnitude and speed of warming) is largely due to the feedback loops that were discussed earlier.14

In my reading of the literature, the assumptions on the damage function are most problematic. Pindyck (2013, p. 868) remarks: “Sometimes these numbers are justified by referring to the IPCC or related summary studies. For example, Nordhaus (2008) points out that the 2007 IPCC report states that "global mean losses could be 1-5 percent GDP for 4°C of warming". But where did the IPCC get those numbers? From its own survey of several IAMs. Yes, it’s a bit circular.15

The bottom line here is that the damage functions used in most IAMs are completely made up, with no theoretical or empirical foundation. … I do not want to give the impression that economists know nothing about the impact of climate change. On the contrary, considerable work has been done on specific aspects of that impact, especially with respect to agriculture. … Statistical studies of this sort will surely improve our knowledge of how climate change might affect the economy, or at least some sectors of the economy. But the data used in these studies are limited to relatively short time periods and small fluctuations in temperature and other weather variables.16 In other words, the damage functions used to estimate the social cost of carbon are either completely arbitrary or need to make considerable out-of-sample predictions.

I still have conflicting feelings about keeping dogs.17 While I do think that Luna is much better off with Alef than that she was before the adoption, and the frequent wiggling of the little stump of a tail suggest that she is indeed happy most of the time, at other moments I feel like we keep her prisoner. I don’t particularly like walking her on a leash, but it seems that we have few other options until we master sign language for dogs. We let her off the leash at the beach, and after swimming, running and digging holes for a half an hour, Luna returns to us. So, last Friday we felt confident enough to let her off the leash in the park, and off Luna went. Alef called señora Rosie to confirm that Luna was with her and señor Eusebio, with whom Luna had been staying while Alef was travelling. And indeed, Luna was with them and could stay for a sleepover.

We took advantage of having the night to ourselves and went to Sayulita for dinner. The waiter at Chili Relleños congratulated us on being the first couple that week holding each other’s hands more than their smartphones. As it was a Friday night, we went to dance in one of the famous beach parties after dinner. Dancing under stars does indeed add something to the experience, as well as having a drink or a smoke in the open air between hermit crabs. At the beach party, we ran into Karlos, and after a quick chat with Alef, he spun her around her axis a few times and handed her hand back to me. We stayed on the dancefloor for a few more songs, and while watching Alef dance against the backdrop of a smoke-filled dancefloor with laser decorations, I congratulate myself once more on my decision to come to Mexico.

We decided that it would be nice to get a new mattress, as the patchwork of foam was starting to take its toll on our backs, but that we first needed a door to our bedroom to prevent Luna from digging holes in the new mattress. Yesterday, we received a triplex plate that was going to serve as a sliding door to our bedroom. When grabbing the triplex plate, I noticed a little scorpion just centimeters away from my right hand, so I gently put the plate down again. I caught it in a plastic cup and called Alef. I admired and feared the little creature for a moment, and regret our decision afterwards. I found watching a scorpion drown in alcohol a truly sad sight, it was trying to defend itself by trying to sting the liquid. It reminded me a little of watching shrimps or lobsters struggle in a pan with boiling water. Next time we should just take far enough from the village and set it free. After seeing Luna playfully remove the legs of a cricket later that evening, I realized that she might be the reason why we didn’t see any scorpions earlier. Keeping a dog also has its advantages.


I realize that my answer to the question what makes for an acceptable Pigouvian tax has not been satisfying. At least, the arbitrary assumptions on damage functions haven’t satisfied me. Once more, Pindyck (2017, p. 349) worded it very clearly: “The difficulty with the use of IAMs for policy analysis goes beyond their arbitrary parameter assumptions and ad hoc damage functions. The greater problem… … is that they create a perception of knowledge and precision that is illusory, and can mislead policymakers into thinking that the forecasts generated by the models have some kind of scientific legitimacy.” However, the lack of consensus on the size of SCC should not hinder implementation of a carbon tax. Settling on a politically acceptable level, “would help to establish that there is a social cost of carbon, and that social cost must be internalized in the prices that consumers and firms pay. … Later, as we learn more about the true size of the SCC, the carbon tax could be increased or decreased accordingly.18

The arguments for a rapid introduction of a Pigouvian tax, reminded me of a similar argument in favor of swift move towards a zero-growth economy: “There are a large number of steady-state levels of stocks to choose from, and such a choice is a difficult problem of ecology and ethics. But our inability to define the optimum level does not mean that we will not someday discover that we have grown beyond it. It is more important to learn to be stable at existing or nearby levels than to know in advance which level is optimal. Knowledge of the latter without the former merely allows us to recognize and wave goodbye to the optimum as we grow through it. Besides, the optimum may well be a broad plateau within which one place is as good as another as long as we don't go too near the edge.19 The puzzle of economics has become increasingly intertwined with ecology and ethics. It leaves me wondering how measure the value nature and value nature in our public decision-making.

Chapter 8



“Contemplation of the world’s disappearing supplies of minerals, forests, and other exhaustible assets had led to demands for regulation of their exploitation. The feeling that these products are now too cheap for the good of future generations, that they are selfishly exploited at too rapid a rate, and that in consequence of their excessive cheapness they are produced and consumed wastefully has given rise to the conservation movement.”

 Harold Hotelling (1931, p. 137)


1. Sanders joined in 2013 and Tillerson in 2009, see Sanders (2013) and Gold and Talley (2009), respectively.

Anchor 1

2.  See Mankiw (2006), who further lists tax incidence, economic growth and national security as benefits of the tax. Pindyck (2017, p. 347) further lists easing international climate negotiations, monitoring, and lower political costs of raising taxes as advantages of the Pigouvian tax as compared to Paris type negotiations over emission reductions. The possibility to decreases taxes elsewhere is known to the literature as the double dividend hypothesis. I return to the double dividend hypothesis in Chapter xx. Given the obesity problems and large meat consumption in many countries, I would be tempted to include health benefits of a meat tax.

3. Daily et al. (2000, p. 395) contains an optimistic example: “Since 1997, the government of Costa Rica has been paying landowners for several ecosystem services: carbon sequestration and protection of watersheds, biodiversity, and scenic beauty. The payments, about U.S.$50/ha-yr, are financed in part by a tax on fossil fuels and are resulting in significant forest conservation and restoration.

4. While Pigou (1920) did not refer to Jevons (1879), he quoted other works liberally. Jevons called it the problem of economics, but in my experience, it is more conducive to invite people to solve puzzles.

5. For economists is common to think in marginal changes, because the solutions to optimization problems usually require that marginal benefits of something are equal to the marginal costs of something else.

6. Jevons even has his own number, it is 8.616.460.799, in case you wonder.

7. In case we’d split the bill I’d have an externality on your budget (and you vice versa on mine), whereas we do not have uncompensated damage on each other’s wallets if we both pay for our own meal; for some reason, the latter is called going Dutch. I’d like to go Dutch on our emissions.

8. A second solution to the problem of externalities is treated in Chapter 11.

9. There is good reason to distinguish between a tax on carbon and a tax on methane. The social cost of methane is likely to be higher; as it is a more potent greenhouse gas and has a shorter half-life in the atmosphere. So, reducing methane emissions reduces global warming faster.

10. For anyone in the same predicament, see for useful tips on how to train deaf dogs.

11. See Pindyck (2017, p. 350). Chapter 13 contains a discussion why using an average SCC might be preferred for policymaking.

12. See Pindyck (2017, p. 345).

13. I replaced the abbreviation GHG in the quote for its completely written out form greenhouse gas. I return to the topic of discounting (i.e. how to account for future costs or benefits in current decisions) in Chapter xx and Chapter 13 contains another discussion of the SCC.

14. See Chapter 5.

15. IPCC refers to International Panel on Climate Change. Or to be more specific on the assumption, see Pindyck (2013, p. 867): “Most IAMs (including the three that were used by the Interagency Working Group to estimate the SCC) relate the temperature increase T to GDP through a "loss function" … For example, the Nordhaus (2008) DICE model uses [an] inverse-quadratic loss function. Weitzman (2009) suggested [an] exponential quadratic loss function, which allows for greater losses when T is large. But remember that neither of these loss functions is based on any economic (or other) theory. Nor are the loss functions that appear in other IAMs. They are just arbitrary functions, made up to describe how GDP goes down when T goes up.” I introduced the brackets [an] to replace ‘the following’, because I omitted the specific functional forms from the quote. See Chapters 5 and 12 for discussions that GDP is a limited measure for social welfare.

16. Pindyck (2013, p. 868) includes references to a growing literature of climate change impact evaluation studies.

17. Keeping dogs is also an often-used example of externalities of consumption. The owner enjoys the dog, but the neighborhood may not enjoy its barking and feces. Some municipalities have a Pigouvian tax on dogs (e.g., the municipality of Den Haag charges a dog tax of €114 for keeping a dog in 2017, €179 for the second, €227 for third or more).

18. Pindyck (2013, p 870). Without the tax it is left up to consumers discretion to offset emissions; e.g. through donations to organizations such as,, or

19. See Daly (1974, p. 16). The steady-state levels of stocks refer to the size of the economy, see also and Chapter xx.  

Anchor 2
Anchor 3
Anchor 4
Anchor 5
Anchor 6
Anchor 7
Anchor 8
Anchor 9
Anchor 10
Anchor 11
Anchor 12
Anchor 13
Anchor 14
Anchor 15
Anchor 16
Anchor 17
Anchor 18
Anchor 19
Anchor 20
bottom of page