- Email: [email protected]
The Carbon Trading Game
Climate Policy 3S2 (2003) S143–S155
The Carbon Trading Game Roger Fouquet∗ Environmental Policy and Management Group, Imperial College London, Prince Consort Road, London SW7 2BP, UK Received 20 April 2003; received in revised form 3 September 2003; accepted 3 September 2003
Abstract In response to the Kyoto Protocol, an international market for carbon dioxide tradable permits is likely to be created. Two of the key issues involved are explaining the concepts of tradable permits to industrialists, policy-makers and the man on the street, and anticipating how the market will evolve. A simple game of the market for carbon dioxide tradable permits has been developed and used that can help deal with both issues. As a pedagogical tool, this game benefits from simplicity (just a few pieces of paper are needed) and enables students to grasp the concepts and remember them through the intensity and fun of a trading ‘pit’. The experiences also provide substantial insights into the evolution of the carbon dioxide permit market, particularly related to the evolution of trade volume, permit prices and country strategies. © 2003 Elsevier B.V. All rights reserved. Keywords: Carbon dioxide permits; Carbon Trading Game; Kyoto Protocol
1. Introduction The market for carbon dioxide permits is in the process of developing from a pipedream of a few economists into a reality with international trades exchanging dollars for ‘carbon credits’ and companies making profits by brokering supply and demand. All this is driven by the ideas outlined in the Kyoto Protocol. For many though, the concept of selling the right to pollute is odd, possibly unethical and certainly confusing. To ensure that more people appreciate the power of tradable permit schemes and even get involved in making the market for carbon dioxide successful, there is a need to familiarise non-academics with the concepts of tradable permits and how they work in practice (Braaddbaart, 1998). With this objective and the pedagogical mantra “tell them and they will forget, show them and they will remember, involve them and they will understand” in mind, about 5 years ago, I developed a game to introduce mostly non-economist students to the concept of tradable permits. This game has proved an effective and rewarding way to convince them of the value, as well as the complexities, of pollution permits. There are ∗
Tel.: +44-207-594-9318; fax: +44-207-594-9304. E-mail address: [email protected] (R. Fouquet). 1469-3062/$ – see front matter © 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.clipol.2003.09.011
S144
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
other, more complex simulations that can be used, say, using computer programmes, which will no doubt replicate more accurately the context of a potential carbon dioxide market. Nevertheless, the game my students used had certain pedagogical advantages: its simplicity (just a few pieces of paper are needed) and little is better for grasping the concepts and remembering them than the intensity and fun of trading in a ‘pit’. The purposes of this paper are, first, to present the Carbon Trading Game to allow lecturers to use the game in seminars either with economic students or with people that may be involved or interested in such a market (industrialists, policy-makers, NGOs) and, second, to show how the markets evolved, when the game is played. This paper should contribute to a better presentation of the concept of tradable permits to students and stakeholders, and will also give us some insights into how a market for carbon dioxide emission permits may develop. In the next section, the Carbon Trading Game is explained. Section three summarises the insights generated from running the game numerous times that might be of value to analysts. The final section makes a few concluding comments.
2. The Carbon Trading Game The Carbon Trading Game allows individuals to act as the leader of a nation involved in buying or selling tradable permits for carbon dioxide emissions (see the Appendix for the exact format of the hand-out given to students). There are 10 countries (including Brazil, Canada, China, the EU, India, Indonesia, Japan, Russia, the Ukraine and the USA) with different costs of reducing emissions. The differing costs lead to gains from trading permits; those that have high costs of abatement will pay for countries with lower costs to reduce their emissions. The objective of the ‘role-playing’ game is to develop a feeling for how economic theory might work in practice: in this case, how individual countries or companies would behave when faced with commitments to the Kyoto Protocol and the possibility of tradable permits to comply with them. The following section describes how the game proceeds: It is recommended that before formally starting the game the lecturer gives the students/players a brief overview of the tradable permit markets (for more details, see for example Tietenberg, 1990). With a theoretical understanding of the concept, the incentives driving traders (especially the role of marginal costs of abatement and prices), the forces leading to an equilibrium quantity and price outcome and the benefits of flexible mechanisms for reducing pollution, the students should be ready to use their knowledge in practice. Each player is given a handout (Appendix A) representing one of the ten countries. There can be one to five players per country, which allows for substantial flexibility in the size of the audience. The handout has explanations of the game on one-side and, on the other, graphs of the country’s marginal costs of abatement and the total costs of abatement. The players’ objective is to save or make as much money as possible while ensuring that their country has sufficient permits for their level of emissions. Players are invited to examine their marginal costs of abatement curve (see Fig. 3 in the Appendix). The information is in graphical rather than numerical form because it is easier to interpret a curve and because it replicates the degree of uncertainty that many countries would be faced with when assessing their costs of abatement. Players are then asked to bid for a tonne of carbon in an auction. Each country secretly submits an estimate of their willingness to pay. The lecturer then collects the bids and presents the bid-prices anonymously on a graph.
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
S145
Table 1 Current and target carbon emissions (in millions of tonnes of carbon) Country
Current emissions
Target emissions (first round)
Target emissions (second round)
Brazil Canada China EU India Indonesia Japan Russia Ukraine United States
100 100 1400 800 1000 100 300 500 100 1400
90 90 1400 700 1000 100 270 500 100 1200
80 80 1260 600 900 90 240 450 90 1000
This acts as a valuable source of information about the range of possible prices. Prior to the auction, the players only had information about their marginal costs of abatement. Now, they can begin to identify (within a range) where the price line meets the marginal cost of abatement curve. For all emissions where the marginal cost of abatement lies below the price line, the player should reduce emissions domestically. Whether these reductions are sold as excess permits or help to avoid buying permits depends on the baseline and target (i.e. the allocated) level of emissions (see Table 1). After the auction, which acts only as a source of information rather than as of collecting permits (although the game could be devised in that way), the trading begins. Players need to wander around the room, asking others how many permits they want to sell/buy and at what price (in US$ per tonne). Once they have assessed some of the offers, they negotiate (buyers trying to drive down the price, sellers trying to push up the price) until they can strike an agreement. All trades need to be registered on a board indicating the seller, the buyer, the quantity traded (in millions of tonnes) and the price. The board acts as vital information about the evolution of prices, enabling players to reassess the quantity of permits they should buy or sell. The trade continues for around 15 min, preferably with a 2 min warning of the end, which often acts as a catalyst for last minute bids. Once all trades are completed, the pit is closed. There is an assumption that all emissions not covered by permits must (and will automatically) be reduced by the country, since this is likely to be cheaper than paying the US$ 100 per tonne fine. So, whether a country is selling permits or just has not bought enough permits, the emissions are reduced through (non-specified) domestic reduction at the marginal cost of abatement and players need do nothing additional for these reductions to occur in the game. The game is then discussed. Special focus should be given to the evolution of prices and the gains from trade. The total cost of abatement curve (Fig. 4 in the Appendix) enables players to assess their situation and discuss their gains. For buyers of permits, Fig. 4 tells them the cost they would have incurred if they had not purchased extra permits. This can be compared with the actual total cost of buying the permits, and a cost–benefit analysis performed. For sellers, Fig. 4 indicates the actual cost of reducing pollution and selling permits. This can be compared with the revenue earned from the sale. Nearly always, buyers have saved money and sellers have made a profit by trading permits. A second round is often valuable to consolidate the understanding and alter the situation. In particular, the lecturer (and market regulator) can propose new post-Kyoto reduction targets. As shown in the last
S146
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
column of Table 1, countries with reduction targets in the first round should double the reductions (e.g. the EU, with an initial target reduction from 800 million to 700 million, now has a target 600 million). Note that countries should start with the same level of reduction as in the initial situation (e.g. the EU starts from 800 million tonnes again). The argument is that, each year emissions are reduced below a baseline, there is a cost incurred; the cost is not a once-and-for-all expenditure, but rather an on-going one. The trading-pit is re-opened for a further 15 min. Trades happen in very much the same way as the first round, except players have a firmer grasp of their objectives and fewer total permits exist—this does not mean less trade though. Once completed, the lecturer should conclude with a final discussion of the game’s objectives, the players’ experiences, the evolution of the market trades and prices, and how changing the situation can alter the market outcome. With more time additional changes to the game can be made: the level of reduction required, the level of information about prices (e.g. making trades a secret or announcing the price in every trade), the number of countries involved (with twenty or thirty countries there would be more trading), the cost of each trade (which is zero in the game played but is positive in reality), the nature of the trading (computer-based trading as in a stock market could increase the amount of trading, although much of the fun of face-to-face interaction would be lost). Lecturers should allocate 90 min for the whole game, including the introductory talk on market permits, the explanation of the game, the development of the strategy, the trading and the conclusion.
3. The experiences 3.1. An overview Having used the game as a pedagogical tool more than a dozen times since its creation in 1998, I have some comments to make about the experience and about the nature of tradable permit markets that may be of interest to tutors and analysts. The game was played in four separate courses, the MSc in environmental engineering and the MSc in energy economics at the University of Surrey, the MSc in environmental technology at Imperial College, and an environmental workshop run for members of the Foreign and Commonwealth Office. All the participants I have had the pleasure of running the game with were, therefore, bright, environmentally aware and (for the most part) non-economists. The predominant mood during the games is initially one of slight confusion, followed by great enthusiasm, interaction and enjoyment. All the students finish the game by understanding the concept of tradable permits, the role of prices and the marginal costs of abatement, and the complexities of the practicalities of running such a scheme. It also seems to leave a more long-lasting memory of the experience, compared with most lectures on the subject. 3.2. Trade volume There is always a tentative beginning to the first trading round. Players are a little uncertain about their strategy, but once the first trade is marked-up on the board, understanding sets in, and generally confirms beliefs. Table 2 presents the trade volume and prices in seven different games, and the averages. In the first round, baseline emissions in the ten countries sum to 5800 million tonnes, but there are only enough
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
S147
Table 2 Trade volume and prices in Carbon Trading Games Game/round
Trade volume (million)
Average price (US$)
Game/round
Trade volume (million)
Average price
1/1 2/1 3/1 4/1 5/1 6/1 7/1 Average Equilibrium
297 154 725 264 356 252 199 321
11.7 22.0 32.4 9.4 26.2 22.6 12.4 19.5 8
1/2 2/2 3/2 4/2 5/2 6/2 7/2 Average Equilibrium
625 450 620 565 410 438 385 499
30.8 47.0 59.2 25.2 75.8 29.5 26.0 41.9 32
permits for 5450 million tonnes of emissions—a reduction of 6% is, therefore, needed. The average level of trade is 321 million, or 6.7% of the total number of permits. There is substantial variation between games, with some under 200 million and some over 700 million. In the second round, the total supply of emissions falls to 4790 million tonnes, a 17.4% reduction in emissions relative to the baseline. The average level of trade volume increased to nearly 500 million, or 9.2% of the total supply of permits. The larger reductions from countries such as the US, EU and Japan, and the high marginal costs of abatement they would incur if they had to reduce all the emissions themselves, creates an increase in the demand for imported permits. 3.3. Prices A brief examination of the prices paid for permits in the games provide useful information about the ranges players’ were willing to pay for permits, and the value of trading. Fig. 1 represents the first round trades, which approximates to the Kyoto Protocol requirements. The lowest any country sold at was US$
Fig. 1. Prices of trades for seven games (round 1—Kyoto-style reductions).
S148
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
Fig. 2. Prices of trades for seven games (round 2—post-Kyoto reductions).
3, and the highest any country bought at was US$ 50. The average price for the seven games was US$ 19.5 per tonne (see Table 2), with the majority of trades being in the US$ 5–20 price range. This is entirely within the range of what is expected from trades related to the Kyoto Protocol. Fig. 2 represents more significant emission reductions. We see that at least one country was willing to pay US$ 100 for permits. It was a trade between the Chinese and the US. That particular game, no. 5, experienced collusion (see below) between China and India, which drove up the prices in both rounds, explaining the average price of US$ 75.8 per tonne. Brazil similarly bought from China at US$ 95 per tonne. In other games, some managed to buy for as low US$ 10 per tonne. Most trades were in the US$ 15–60 price range. This second round of trade shows the impact of reducing supply on prices. A 12.1% decline in supply has more than doubled average prices. 3.4. Price evolution An interesting issue to comment on is the evolution of prices. Figs. 1 and 2 show how prices diverge from the level at which all countries’ marginal costs of abatement equalise (as theory would suggest). The theoretical equilibrating prices are US$ 8 per tonne in the first round (i.e. for a reduction of 350 million tonnes) and US$ 32 per tonne in the second round (i.e. for a reduction of 1010 million tonnes). Instead, they depend on players’ understanding of and strategy within the game, and how these change with experience. The first trade often acts as an anchor, determining the price of future trades. For example, in game 1, India sold 10 million tonnes to Canada at US$ 10. In that round, the highest price paid for a permit was US$ 20. At the other end of the scale, India and the EU completed the first trade of game 3 at US$ 25 per tonne. The bids in that round ranged from US$ 17 to 50. And, the anchoring appears to have long-term effects, as it influences the next round too. Game 1 prices ranged from US$ 10 to 40, whereas game 3’s varied from US$ 40 to 70. Another issue of importance is the final flurry of activity. Often, especially in the first round, minimal trade takes place until the two-minute warning. At that point, players realise they need to buy permits, and rush into deals, often driving up the price substantially. Both rounds of game 5 reflect this well. Because
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
S149
of the strong supply side, few trades had been made. Then, with time running out and the need to buy pressing, prices spiralled. In round 1, the average price was US$ 26.2, but the last three trades were US$ 40, 40 and 50. In round 2, although high prices had been agreed before, they had started dropping; but, when the end of the round was signalled, the last three trades drove prices up again, to US$ 60, 70 and 100. The opposite reaction can also take place. Selling countries realise that they would lose out by not completing a few last deals. So, after playing a hard-bargaining game, they become desperate to sell to anybody. In the first round of game 4, India was willing to sell to the EU for US$ 3 per tonne. The last few trades of games 1–4 were all tending downwards, as countries with low marginal costs were selling on the cheap, quickly. Such last-minute behaviour is similar to what happened in 1999 just before the second, more expensive phase of the US sulphur dioxide permit scheme (see US EPA, 2003) and might occur in 2012 for carbon dioxide permits. 3.5. The key players The total emissions and marginal costs are crucial for the power a country can wield in the market. The big players, in the game, are generally China and India on the supply side and the EU and the USA on the demand side. Ukraine and Indonesia are clearly small fish in a big pond. They often feel frustrated by their inability to sell more than a few million tonnes, which limits their interaction to a few smaller buyers. As mentioned before, a strategy developed in game 5 can highlight the potential for market power. The Chinese refused to trade in the first round. Initially, this seemed a folly: the country would lose millions in trade revenue. In this round, prices were (probably) higher (averaging at US$ 26.2 per tonne) than if China had traded. On the whole, countries with excess permits, principally poorer countries with low marginal costs of abatement, especially India, sold at relatively high prices. In the next round, target reductions were even higher; if China continued its strategy, prices would sore dramatically. When the round started, one of the Chinese representatives suggested it would hold out, and prices did sore. China then entered the market, selling in two deals one at US$ 95 and the other at US$ 100 per tonne. The gamble paid off, and China made a net profit of around US$ 8 billion from those two trades—perhaps more than if they had joined in from the beginning. The strategy also aided other low-income countries, at the expense of industrialised ones needing permits. On the demand side, Japan is a highly sought after customer, because of its very high marginal costs of abatement. This is especially true in the early trades, until its representatives have gathered information about converging prices. In addition, as mentioned earlier, early trades can influence the evolution of the market. In such a market, where Japanese representatives are not approaching faceless individuals but rather reliable and trusted suppliers, they may seek to develop long-term trade relationships with certain other countries. The degree to which early trades, and associated long term relationships (along with features such as trust and certainty of supply), are influential will vary according to personal traits of representatives and cultural aspects of different trading countries. In this context, suppliers (especially small ones) can benefit greatly from developing early trade relationships with countries experiencing high marginal costs of abatement. 3.6. Savings and profits The main purpose of tradable permits is to minimise the cost of achieving total emission targets. For countries with reduction targets and high costs of abatement, they can save money by paying another
S150
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
country to take responsibility for reducing emissions. And, the countries taking responsibility for reductions earn considerable (often desperately needed) foreign revenue. The Carbon Trading Game shows this principal to students quite effectively. Almost all players saved or earned money. The only ones that failed to gain from the trades were those that miscalculated. They sold or bought too many permits, and in the dying seconds of the rounds could be seen running around desperately trying to buy or sell back some permits. Even they generally broke even. There is considerable difference in the sums saved or earned though. The big players tended to gain the most from trading. The Japanese saved several billion dollars, especially when the target reductions were increased in the second round. The Chinese and the Indians compete for the highest earnings, although China should win because of its larger quota of permits. In the first round, they can expect to earn around US$ 2–3 billion, and up to US$ 5 billion in the second round. Countries in the middle range (i.e. low marginal costs for a buyer, high marginal costs for a seller) have the least to gain from the trading, and this includes the US (See Ellerman et al., 1998 for more analysis). The gains from trade vary considerably with the nature of the market. Inevitably, when prices were high, as in games 3 and 5, the sellers were earning very large profits. When prices low, like in games 1, 4 and 7, their profits were more modest and the buyers saved most. 3.7. The role of individual characteristics and information Outcomes also varied according to the players involved, which will be touched on next. More generally, the characteristics of individual players (and the group dynamics) can have a great influence on the market and its evolution. Certain traders are more willing to take risks or drive hard bargains - this was reflected in the market price. Extraverted players are more likely to wander between country representatives, comparing different offers, thus having a better ability to find the best deal. They are particularly important for initiating debate over prices and quantities, and may also be more willing to propose alternatives when disagreements appear. Nevertheless, the strong, silent type can drive a hard bargain by his/her appearance of not shifting from the offer. Thus, both extraverted and introverted players (or countries) can bargain to their strengths, making for a potentially interesting outcome. Crucial to the success of these markets is an adequate supply of information. After analysing the marginal cost of abatement curve, the auction round provides information on the range of prices that might be paid. It is played with a degree of confusion, as some players are still not yet clear of their objectives and how to achieve them. The board acts as a vital source of information. It tells players what trades have taken place, which countries are buyers and which are sellers, and the current price for a permit. Producing a graph on the board as the trades take place recorded can be effective for helping players detect trends. In a couple games, I experimented with the supply of information. When each trade was made, the price was called out. We experienced substantially more activity and more stable prices than when information was less freely available, even though the only cost of the information to players was to go to the front of the classroom and examine the board. Just like real markets, there are other ways of gathering information. Espionage, sneaking over other teams’ shoulders, is popular and can help gather information about abatement curves, need for permits and prices willing to concede. Also, the players can consult outsiders (in this case, the lecturer) for advice on whether to agree a trade. No consultancy fees were paid.
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
S151
4. Conclusion This paper introduces the Carbon Trading Game and some of the observation from running it with students. The objective of the ‘role-playing’ game is to develop a feeling for how individual countries (and, thus, their companies) would behave when faced with commitments to the Kyoto Protocol and the possibility of tradable permits to comply with them. The game allows students to act as the leader of a nation involved in buying or selling tradable permits for carbon dioxide emissions to ensure they match emissions with permits at least cost. There are 10 countries (including Brazil, Canada, China, the EU, India, Indonesia, Japan, Russia, the Ukraine and the USA) with different costs of reducing emissions. The students are presented with their country’s marginal costs of abatement curves and total abatement curves. Students then need to collect information about potential prices per tonne to assess the trades (buying or selling permits) they will pursue. The game works very well at teaching the value of tradable permit schemes and some of their complexities, as well as being lots of fun. Another purpose of presenting this game is as an exercise in observing the nature and evolution of tradable permit markets for carbon dioxide. Each game generates interesting information about the nature of the market. The level of trade and prices vary substantially between games. The overall average trade volume was 321 million tonnes (or 6.7% of the total supply of permits) at an average price of US$ 19.5 per tonne of carbon for a total 6% Kyoto-style reduction target, and 499 million tonnes (or 9.2% of the total supply of permits) at an average price of US$ 41.9 per tonne of carbon for a 19.4% post-Kyoto-style reduction target. Running such experiments can also provide valuable lessons about the evolution of prices that might not be anticipated from economic theory. For example, the price agreed in the first trade has major implications for future trades. While there may be some adjustment towards the equilibrium price, the whole game is thrown in a particular direction. Also, the final flurry of trades often leads to an upward or downward spiral in prices. In addition, the evolution of prices depends crucially on the role of large players using their market power (Malik, 2002) and of technological innovations (Montero, 2002). Another important lesson is that this international permit system does appear to save industrialised countries money and earn developing nations highly sought-after foreign revenue (Stevens and Rose, 2002). While over-dependence on a market that could collapse—if, say, new technologies provide cheap ways of reducing carbon dioxide emissions—would be careless, these experiments remind us that many countries can gain from the creation of a market where the commodity traded is the responsibility for reducing pollution. Clearly, a simplified game representing a tradable permit scheme cannot do justice to the complexities of a real international market. For reviews of the market and institutional developments associated with an international tradable permits for carbon dioxide see Baron (2001) and Christiansen and Wettestad (2003). Nevertheless, the lessons from the games played can be of relevance to real carbon permit markets. An understanding of the nature of different countries (and their companies)—whether they are major or minor players, pro-active or reactive, informed or confused—will have important implications for how they interact and trade. The Carbon Trading Game will hopefully help many become more aware of potential gains from being involved in the burgeoning international market for tradable permits. It may also help some more vulnerable groups avoid the pitfalls of an international market for the right to emit carbon dioxide.
S152
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
Acknowledgements I would like to thank many enthusiastic students who participated in the game (mostly from the Imperial College MSc in environmental technology, the University of Surrey MSc in environmental engineering and members of the Foreign and Commonwealth Office) and Martin Hession, Matt Leach, Peter Pearson, Tanya O’Garra, Suzy Hodgson and Tim Jackson for their encouragement in developing and drumming-up interest in the game. I am also grateful to Denny Ellerman, Hank Jacoby and Annelene Decaux for allowing me to use their data for the purpose of the game. Appendix A. Example of the game The following is an example of the hand-out for country A:
A.1. Tradable permits game (country A): Students will have an opportunity to test their understanding of how tradable permits systems achieve the desired level of pollution and reduce the costs of abating pollution. They will play a game simulating a worldwide tradable permits market for carbon dioxide emissions. The following text would be form the background for the person (or persons) representing country A.
A.2. Context of the discussion: You (and your colleague) are representing country A. Your current carbon dioxide emissions are at current emissions 100 million tonnes per year and you have agreed to reduce them to 90. This means that you have 90 permits each allowing you to pollute 1 million tonnes Fig. 3. A tradable permit scheme has been agreed upon, which means you can buy a number of permits from other countries at a price in US$ per permit (such that you are entitled to pollute an additional amount equal to the number of permits) or you can sell a number of permits to other countries (such that you are entitled to emit carbon dioxide according to your new amount). Anyone failing to comply with emissions permits will be fined US$ 200 million per 1 million tonnes over your permitted amount. With this agreement in mind and the knowledge that you have recently been elected, you are trying to do ‘what is best for the country’.
Brazil
300
$ per tonne
Marginal Cost of Abatement
400
200
100
0
76
80
84
88
92
CO2 Emissions Fig. 3. Marginal cost of abatement.
96
100
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
S153
Brazil
3000
Million $
Total Cost of Abatement
4000
2000
1000
0
76
80
84
88
92
96
100
CO2 Emissions Fig. 4. Total cost of abatement. Source: figures based on Fig. A.4 of A. Denny Ellerman, Henry D. Jacoby and Annelene Decaux (1998) The effects on developing countries of the Kyoto Protocol and CO2 emissions trading. Working Paper 41. Joint Program on the Science and Policy of Global Change. MIT (available on webpage: http://www.web.mit.edu/globalchange/www/rpt41.html).
A.3. Your objective: Your objective is to buy or sell emissions with other countries in an attempt to keep the costs of reducing emissions as low as possible. At the end of each year, we will assess your actual total costs of reducing emissions (based on the emissions reduced, the costs of reducing these emissions, the earnings from selling permits and the costs of buying permits) relative to your initial total costs of abatement (represented in Fig. 4).
A.4. Trading: 1. In Year 1, to start the game, and to provide information about prices, an auction will take place, where permits will be allocated to the four highest bidders. Please write how much your country is willing to pay per permit. The four highest bidders will receive 2% of their current emissions. Thus, paying that price for 2% and increasing permitted emissions by 2%. 2. Calculate the total cost of abatement, according to your new or unchanged permit level (based on the emissions reduced, the cost of reducing these emissions, the earnings from selling permits and the costs of buying permits). 3. The total costs of abatement will be written in a log book. 4. In subsequent years, there will be no auction, just trading. 5. You have X minutes, to agree trades with other countries. You can agree as many sales or buys as you want, provided you have permits available to sell. 6. When a trade has been agreed, the two countries need to write in the long book the quantity of permits and the price per permit traded, and the countries involved. 7. Calculate the total cost of abatement, according to your new or unchanged permit level (based on the emissions reduced, the cost of reducing these emissions, the earnings from selling permits and the costs of buying permits).
Appendix B. Data for the countries The marginal costs of abatement for the 10 countries were approximated from the equation provided in Table A.1 of Ellerman et al. (1998). The total cost is calculated as the sum of all the marginal costs. For example, the total cost for Brazil reducing emissions down to 94 is equal to the sum of the marginal cost at 98 million tons (i.e. US$ 20/tonne) times two million, at 96 million tonnes (i.e. US$ 40/tonne) times two million and at 94 million tonnes (i.e. US$ 60/tonne) times two million; 2 million × 20 + 2 million × 40 + 2 million × 60 = US$ 240 millions. Note that the units of measurement are metric tons of carbon.
S154
Table A.1 Marginal and total costs of abatement at specified levels of emissions (in million tons of carbon) for the 10 countries in the Carbon Trading Game Brazil Emission
TCA (million $/tonne)
400 340 280 220 180 150 120 100 80 60 40 20 0
3980 3180 2500 1940 1500 1140 840 600 400 240 120 40 0
60 56 52 48 44 40 36 32 28 24 20 16 13 10 7 5 4 3 2 1 0
14028 12348 10780 9324 7980 6748 5628 4620 3724 2940 2268 1708 1260 896 616 420 280 168 84 28 0
China 840 868 896 924 952 980 1008 1036 1064 1092 1120 1148 1176 1204 1232 1260 1288 1316 1344 1372 1400
Emission 70 72 74 76 78 10 82 84 86 88 90 92 94 96 98 100
EU MCA (US$/tonne)
TCA (million $/tonne)
300 240 200 160 130 100 80 70 60 50 40 30 20 10 5 0
2990 2390 1910 1510 1190 930 730 570 430 310 210 130 70 30 10 0
90 80 70 60 55 50 42 36 30 25 20 16 13 10 7 5 4 3 2 1 0
12380 10580 8980 7580 6330 5280 4280 3440 2720 2120 1620 1220 900 640 440 300 200 120 60 20 0
India 600 620 640 660 680 700 720 740 760 780 800 820 840 860 880 900 920 940 960 980 1000
Japan
Emission
MCA ($/tonne)
TCA (million $/tonne)
560 576 592 608 624 640 656 672 688 704 720 736 752 768 784 800
400 320 260 220 180 140 120 100 80 60 50 40 30 20 10 0
32480 26080 20960 16800 13280 10400 8160 6240 4640 3360 2400 1600 960 480 160 0
150 125 110 100 90 80 70 60 50 40 30 26 22 18 88 10 8 6 4 2 0
2030 1730 1480 1260 1060 880 720 580 460 360 280 220 168 124 14 60 40 24 12 4 0
Indonesia 60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 100
Emission
234 240 246 252 258 264 270 276 282 288 294 300
US MCA (US$/tonne)
TCA (million $/tonne)
400 340 260 200 160 120 100 80 60 40 20 0
10680 8210 6240 4680 3430 2520 1800 1200 720 360 120 0
200 180 160 140 125 110 95 80 70 60 50 45 40 35 30 25 20 15 10 5 0
14950 12950 11150 9550 8150 6900 5800 4850 4050 3350 2750 2250 1800 1400 1050 750 500 300 150 50 0
Russia 300 310 320 330 340 350 360 370 380 390 400 410 420 430 440 450 460 470 430 490 500
Emission
MCA (US$/tonne)
TCA (million $/tonne)
980 1008 1036 1064 1092 1120 1148 1176 1204 1232 1260 1288 1316 1344 1372 1400
200 160 120 100 80 60 50 40 35 30 25 20 15 10 5 0
26600 21000 16520 13160 10360 8120 6440 5040 3920 2940 2100 1400 840 420 140 0
220 200 180 160 140 120 100 80 70 60 50 45 40 35 30 25 20 15 10 5 0
3210 2770 2370 2010 1690 1410 1170 970 810 670 550 450 360 280 210 150 100 60 30 10 0
Ukraine 60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 100
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
76 78 80 82 84 86 88 90 92 94 96 98 100
Canada MCA (US$/tonne)
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
S155
References Baron, R., 2001. International emissions trading: from concept to reality. International Energy Agency, Paris. Braaddbaart, O., 1998. American bias in environmental economics: industrial pollution abatement and incentives versus regulation. Environ. Politics 7 (2), 134–152. Christiansen, A.C., Wettestad, J., 2003. The EU as a frontrunner on greenhouse gas emissions trading: how did it happen and will the EU succeed? Climate Policy 3 (1), 3–18. Ellerman, A.D., Jacoby, H.D., Decaux, A., 1998. The effects on developing countries of the Kyoto Protocol and CO2 emissions trading. Working Paper 41. Joint Program on the Science and Policy of Global Change, MIT. (available on: http://web.mit.edu/globalchange/www/rpt41.html). Malik, A.S., 2002. Further results on permit markets with market power and cheating. J. Environ. Econ. Manag. 44 (3), 371–390. Montero, J.P., 2002. Permits, Standards, and technology innovation. J. Environ. Econ. Manag. 44 (1), 23–44. Stevens, B., Rose, A., 2002. A dynamic analysis of the marketable permits approach to global warming policy: a comparison of spatial and temporal flexibility. J. Environ. Econ. Manage. 44 (1), 45–69. Tietenberg, T.H., 1990. Economic instruments for environmental regulation. Oxford Rev. Econ. Policy 6 (1), 17–33. US EPA, 2003. ‘Monthly average price of sulfur dioxide allowances.’ (available on: http://www.epa.gov/AIRMARKET/trading/ so2market/prices.html).
Further reading Schmalensee, R., Joskow, P.L., Ellerman, A.D., Montero, J.P., Bailey, E.M., 1998. An interim evaluation of the sulfur dioxide emissions trading. J. Econ. Perspect. 12 (3), 53–68. Stavins, R.N., 1998. What can we learn from the grand policy experiment? Lessons from SO2 allowance trading. J. Econ. Perspect. 12 (3), 69–88.
The Carbon Trading Game Roger Fouquet∗ Environmental Policy and Management Group, Imperial College London, Prince Consort Road, London SW7 2BP, UK Received 20 April 2003; received in revised form 3 September 2003; accepted 3 September 2003
Abstract In response to the Kyoto Protocol, an international market for carbon dioxide tradable permits is likely to be created. Two of the key issues involved are explaining the concepts of tradable permits to industrialists, policy-makers and the man on the street, and anticipating how the market will evolve. A simple game of the market for carbon dioxide tradable permits has been developed and used that can help deal with both issues. As a pedagogical tool, this game benefits from simplicity (just a few pieces of paper are needed) and enables students to grasp the concepts and remember them through the intensity and fun of a trading ‘pit’. The experiences also provide substantial insights into the evolution of the carbon dioxide permit market, particularly related to the evolution of trade volume, permit prices and country strategies. © 2003 Elsevier B.V. All rights reserved. Keywords: Carbon dioxide permits; Carbon Trading Game; Kyoto Protocol
1. Introduction The market for carbon dioxide permits is in the process of developing from a pipedream of a few economists into a reality with international trades exchanging dollars for ‘carbon credits’ and companies making profits by brokering supply and demand. All this is driven by the ideas outlined in the Kyoto Protocol. For many though, the concept of selling the right to pollute is odd, possibly unethical and certainly confusing. To ensure that more people appreciate the power of tradable permit schemes and even get involved in making the market for carbon dioxide successful, there is a need to familiarise non-academics with the concepts of tradable permits and how they work in practice (Braaddbaart, 1998). With this objective and the pedagogical mantra “tell them and they will forget, show them and they will remember, involve them and they will understand” in mind, about 5 years ago, I developed a game to introduce mostly non-economist students to the concept of tradable permits. This game has proved an effective and rewarding way to convince them of the value, as well as the complexities, of pollution permits. There are ∗
Tel.: +44-207-594-9318; fax: +44-207-594-9304. E-mail address: [email protected] (R. Fouquet). 1469-3062/$ – see front matter © 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.clipol.2003.09.011
S144
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
other, more complex simulations that can be used, say, using computer programmes, which will no doubt replicate more accurately the context of a potential carbon dioxide market. Nevertheless, the game my students used had certain pedagogical advantages: its simplicity (just a few pieces of paper are needed) and little is better for grasping the concepts and remembering them than the intensity and fun of trading in a ‘pit’. The purposes of this paper are, first, to present the Carbon Trading Game to allow lecturers to use the game in seminars either with economic students or with people that may be involved or interested in such a market (industrialists, policy-makers, NGOs) and, second, to show how the markets evolved, when the game is played. This paper should contribute to a better presentation of the concept of tradable permits to students and stakeholders, and will also give us some insights into how a market for carbon dioxide emission permits may develop. In the next section, the Carbon Trading Game is explained. Section three summarises the insights generated from running the game numerous times that might be of value to analysts. The final section makes a few concluding comments.
2. The Carbon Trading Game The Carbon Trading Game allows individuals to act as the leader of a nation involved in buying or selling tradable permits for carbon dioxide emissions (see the Appendix for the exact format of the hand-out given to students). There are 10 countries (including Brazil, Canada, China, the EU, India, Indonesia, Japan, Russia, the Ukraine and the USA) with different costs of reducing emissions. The differing costs lead to gains from trading permits; those that have high costs of abatement will pay for countries with lower costs to reduce their emissions. The objective of the ‘role-playing’ game is to develop a feeling for how economic theory might work in practice: in this case, how individual countries or companies would behave when faced with commitments to the Kyoto Protocol and the possibility of tradable permits to comply with them. The following section describes how the game proceeds: It is recommended that before formally starting the game the lecturer gives the students/players a brief overview of the tradable permit markets (for more details, see for example Tietenberg, 1990). With a theoretical understanding of the concept, the incentives driving traders (especially the role of marginal costs of abatement and prices), the forces leading to an equilibrium quantity and price outcome and the benefits of flexible mechanisms for reducing pollution, the students should be ready to use their knowledge in practice. Each player is given a handout (Appendix A) representing one of the ten countries. There can be one to five players per country, which allows for substantial flexibility in the size of the audience. The handout has explanations of the game on one-side and, on the other, graphs of the country’s marginal costs of abatement and the total costs of abatement. The players’ objective is to save or make as much money as possible while ensuring that their country has sufficient permits for their level of emissions. Players are invited to examine their marginal costs of abatement curve (see Fig. 3 in the Appendix). The information is in graphical rather than numerical form because it is easier to interpret a curve and because it replicates the degree of uncertainty that many countries would be faced with when assessing their costs of abatement. Players are then asked to bid for a tonne of carbon in an auction. Each country secretly submits an estimate of their willingness to pay. The lecturer then collects the bids and presents the bid-prices anonymously on a graph.
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
S145
Table 1 Current and target carbon emissions (in millions of tonnes of carbon) Country
Current emissions
Target emissions (first round)
Target emissions (second round)
Brazil Canada China EU India Indonesia Japan Russia Ukraine United States
100 100 1400 800 1000 100 300 500 100 1400
90 90 1400 700 1000 100 270 500 100 1200
80 80 1260 600 900 90 240 450 90 1000
This acts as a valuable source of information about the range of possible prices. Prior to the auction, the players only had information about their marginal costs of abatement. Now, they can begin to identify (within a range) where the price line meets the marginal cost of abatement curve. For all emissions where the marginal cost of abatement lies below the price line, the player should reduce emissions domestically. Whether these reductions are sold as excess permits or help to avoid buying permits depends on the baseline and target (i.e. the allocated) level of emissions (see Table 1). After the auction, which acts only as a source of information rather than as of collecting permits (although the game could be devised in that way), the trading begins. Players need to wander around the room, asking others how many permits they want to sell/buy and at what price (in US$ per tonne). Once they have assessed some of the offers, they negotiate (buyers trying to drive down the price, sellers trying to push up the price) until they can strike an agreement. All trades need to be registered on a board indicating the seller, the buyer, the quantity traded (in millions of tonnes) and the price. The board acts as vital information about the evolution of prices, enabling players to reassess the quantity of permits they should buy or sell. The trade continues for around 15 min, preferably with a 2 min warning of the end, which often acts as a catalyst for last minute bids. Once all trades are completed, the pit is closed. There is an assumption that all emissions not covered by permits must (and will automatically) be reduced by the country, since this is likely to be cheaper than paying the US$ 100 per tonne fine. So, whether a country is selling permits or just has not bought enough permits, the emissions are reduced through (non-specified) domestic reduction at the marginal cost of abatement and players need do nothing additional for these reductions to occur in the game. The game is then discussed. Special focus should be given to the evolution of prices and the gains from trade. The total cost of abatement curve (Fig. 4 in the Appendix) enables players to assess their situation and discuss their gains. For buyers of permits, Fig. 4 tells them the cost they would have incurred if they had not purchased extra permits. This can be compared with the actual total cost of buying the permits, and a cost–benefit analysis performed. For sellers, Fig. 4 indicates the actual cost of reducing pollution and selling permits. This can be compared with the revenue earned from the sale. Nearly always, buyers have saved money and sellers have made a profit by trading permits. A second round is often valuable to consolidate the understanding and alter the situation. In particular, the lecturer (and market regulator) can propose new post-Kyoto reduction targets. As shown in the last
S146
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
column of Table 1, countries with reduction targets in the first round should double the reductions (e.g. the EU, with an initial target reduction from 800 million to 700 million, now has a target 600 million). Note that countries should start with the same level of reduction as in the initial situation (e.g. the EU starts from 800 million tonnes again). The argument is that, each year emissions are reduced below a baseline, there is a cost incurred; the cost is not a once-and-for-all expenditure, but rather an on-going one. The trading-pit is re-opened for a further 15 min. Trades happen in very much the same way as the first round, except players have a firmer grasp of their objectives and fewer total permits exist—this does not mean less trade though. Once completed, the lecturer should conclude with a final discussion of the game’s objectives, the players’ experiences, the evolution of the market trades and prices, and how changing the situation can alter the market outcome. With more time additional changes to the game can be made: the level of reduction required, the level of information about prices (e.g. making trades a secret or announcing the price in every trade), the number of countries involved (with twenty or thirty countries there would be more trading), the cost of each trade (which is zero in the game played but is positive in reality), the nature of the trading (computer-based trading as in a stock market could increase the amount of trading, although much of the fun of face-to-face interaction would be lost). Lecturers should allocate 90 min for the whole game, including the introductory talk on market permits, the explanation of the game, the development of the strategy, the trading and the conclusion.
3. The experiences 3.1. An overview Having used the game as a pedagogical tool more than a dozen times since its creation in 1998, I have some comments to make about the experience and about the nature of tradable permit markets that may be of interest to tutors and analysts. The game was played in four separate courses, the MSc in environmental engineering and the MSc in energy economics at the University of Surrey, the MSc in environmental technology at Imperial College, and an environmental workshop run for members of the Foreign and Commonwealth Office. All the participants I have had the pleasure of running the game with were, therefore, bright, environmentally aware and (for the most part) non-economists. The predominant mood during the games is initially one of slight confusion, followed by great enthusiasm, interaction and enjoyment. All the students finish the game by understanding the concept of tradable permits, the role of prices and the marginal costs of abatement, and the complexities of the practicalities of running such a scheme. It also seems to leave a more long-lasting memory of the experience, compared with most lectures on the subject. 3.2. Trade volume There is always a tentative beginning to the first trading round. Players are a little uncertain about their strategy, but once the first trade is marked-up on the board, understanding sets in, and generally confirms beliefs. Table 2 presents the trade volume and prices in seven different games, and the averages. In the first round, baseline emissions in the ten countries sum to 5800 million tonnes, but there are only enough
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
S147
Table 2 Trade volume and prices in Carbon Trading Games Game/round
Trade volume (million)
Average price (US$)
Game/round
Trade volume (million)
Average price
1/1 2/1 3/1 4/1 5/1 6/1 7/1 Average Equilibrium
297 154 725 264 356 252 199 321
11.7 22.0 32.4 9.4 26.2 22.6 12.4 19.5 8
1/2 2/2 3/2 4/2 5/2 6/2 7/2 Average Equilibrium
625 450 620 565 410 438 385 499
30.8 47.0 59.2 25.2 75.8 29.5 26.0 41.9 32
permits for 5450 million tonnes of emissions—a reduction of 6% is, therefore, needed. The average level of trade is 321 million, or 6.7% of the total number of permits. There is substantial variation between games, with some under 200 million and some over 700 million. In the second round, the total supply of emissions falls to 4790 million tonnes, a 17.4% reduction in emissions relative to the baseline. The average level of trade volume increased to nearly 500 million, or 9.2% of the total supply of permits. The larger reductions from countries such as the US, EU and Japan, and the high marginal costs of abatement they would incur if they had to reduce all the emissions themselves, creates an increase in the demand for imported permits. 3.3. Prices A brief examination of the prices paid for permits in the games provide useful information about the ranges players’ were willing to pay for permits, and the value of trading. Fig. 1 represents the first round trades, which approximates to the Kyoto Protocol requirements. The lowest any country sold at was US$
Fig. 1. Prices of trades for seven games (round 1—Kyoto-style reductions).
S148
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
Fig. 2. Prices of trades for seven games (round 2—post-Kyoto reductions).
3, and the highest any country bought at was US$ 50. The average price for the seven games was US$ 19.5 per tonne (see Table 2), with the majority of trades being in the US$ 5–20 price range. This is entirely within the range of what is expected from trades related to the Kyoto Protocol. Fig. 2 represents more significant emission reductions. We see that at least one country was willing to pay US$ 100 for permits. It was a trade between the Chinese and the US. That particular game, no. 5, experienced collusion (see below) between China and India, which drove up the prices in both rounds, explaining the average price of US$ 75.8 per tonne. Brazil similarly bought from China at US$ 95 per tonne. In other games, some managed to buy for as low US$ 10 per tonne. Most trades were in the US$ 15–60 price range. This second round of trade shows the impact of reducing supply on prices. A 12.1% decline in supply has more than doubled average prices. 3.4. Price evolution An interesting issue to comment on is the evolution of prices. Figs. 1 and 2 show how prices diverge from the level at which all countries’ marginal costs of abatement equalise (as theory would suggest). The theoretical equilibrating prices are US$ 8 per tonne in the first round (i.e. for a reduction of 350 million tonnes) and US$ 32 per tonne in the second round (i.e. for a reduction of 1010 million tonnes). Instead, they depend on players’ understanding of and strategy within the game, and how these change with experience. The first trade often acts as an anchor, determining the price of future trades. For example, in game 1, India sold 10 million tonnes to Canada at US$ 10. In that round, the highest price paid for a permit was US$ 20. At the other end of the scale, India and the EU completed the first trade of game 3 at US$ 25 per tonne. The bids in that round ranged from US$ 17 to 50. And, the anchoring appears to have long-term effects, as it influences the next round too. Game 1 prices ranged from US$ 10 to 40, whereas game 3’s varied from US$ 40 to 70. Another issue of importance is the final flurry of activity. Often, especially in the first round, minimal trade takes place until the two-minute warning. At that point, players realise they need to buy permits, and rush into deals, often driving up the price substantially. Both rounds of game 5 reflect this well. Because
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
S149
of the strong supply side, few trades had been made. Then, with time running out and the need to buy pressing, prices spiralled. In round 1, the average price was US$ 26.2, but the last three trades were US$ 40, 40 and 50. In round 2, although high prices had been agreed before, they had started dropping; but, when the end of the round was signalled, the last three trades drove prices up again, to US$ 60, 70 and 100. The opposite reaction can also take place. Selling countries realise that they would lose out by not completing a few last deals. So, after playing a hard-bargaining game, they become desperate to sell to anybody. In the first round of game 4, India was willing to sell to the EU for US$ 3 per tonne. The last few trades of games 1–4 were all tending downwards, as countries with low marginal costs were selling on the cheap, quickly. Such last-minute behaviour is similar to what happened in 1999 just before the second, more expensive phase of the US sulphur dioxide permit scheme (see US EPA, 2003) and might occur in 2012 for carbon dioxide permits. 3.5. The key players The total emissions and marginal costs are crucial for the power a country can wield in the market. The big players, in the game, are generally China and India on the supply side and the EU and the USA on the demand side. Ukraine and Indonesia are clearly small fish in a big pond. They often feel frustrated by their inability to sell more than a few million tonnes, which limits their interaction to a few smaller buyers. As mentioned before, a strategy developed in game 5 can highlight the potential for market power. The Chinese refused to trade in the first round. Initially, this seemed a folly: the country would lose millions in trade revenue. In this round, prices were (probably) higher (averaging at US$ 26.2 per tonne) than if China had traded. On the whole, countries with excess permits, principally poorer countries with low marginal costs of abatement, especially India, sold at relatively high prices. In the next round, target reductions were even higher; if China continued its strategy, prices would sore dramatically. When the round started, one of the Chinese representatives suggested it would hold out, and prices did sore. China then entered the market, selling in two deals one at US$ 95 and the other at US$ 100 per tonne. The gamble paid off, and China made a net profit of around US$ 8 billion from those two trades—perhaps more than if they had joined in from the beginning. The strategy also aided other low-income countries, at the expense of industrialised ones needing permits. On the demand side, Japan is a highly sought after customer, because of its very high marginal costs of abatement. This is especially true in the early trades, until its representatives have gathered information about converging prices. In addition, as mentioned earlier, early trades can influence the evolution of the market. In such a market, where Japanese representatives are not approaching faceless individuals but rather reliable and trusted suppliers, they may seek to develop long-term trade relationships with certain other countries. The degree to which early trades, and associated long term relationships (along with features such as trust and certainty of supply), are influential will vary according to personal traits of representatives and cultural aspects of different trading countries. In this context, suppliers (especially small ones) can benefit greatly from developing early trade relationships with countries experiencing high marginal costs of abatement. 3.6. Savings and profits The main purpose of tradable permits is to minimise the cost of achieving total emission targets. For countries with reduction targets and high costs of abatement, they can save money by paying another
S150
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
country to take responsibility for reducing emissions. And, the countries taking responsibility for reductions earn considerable (often desperately needed) foreign revenue. The Carbon Trading Game shows this principal to students quite effectively. Almost all players saved or earned money. The only ones that failed to gain from the trades were those that miscalculated. They sold or bought too many permits, and in the dying seconds of the rounds could be seen running around desperately trying to buy or sell back some permits. Even they generally broke even. There is considerable difference in the sums saved or earned though. The big players tended to gain the most from trading. The Japanese saved several billion dollars, especially when the target reductions were increased in the second round. The Chinese and the Indians compete for the highest earnings, although China should win because of its larger quota of permits. In the first round, they can expect to earn around US$ 2–3 billion, and up to US$ 5 billion in the second round. Countries in the middle range (i.e. low marginal costs for a buyer, high marginal costs for a seller) have the least to gain from the trading, and this includes the US (See Ellerman et al., 1998 for more analysis). The gains from trade vary considerably with the nature of the market. Inevitably, when prices were high, as in games 3 and 5, the sellers were earning very large profits. When prices low, like in games 1, 4 and 7, their profits were more modest and the buyers saved most. 3.7. The role of individual characteristics and information Outcomes also varied according to the players involved, which will be touched on next. More generally, the characteristics of individual players (and the group dynamics) can have a great influence on the market and its evolution. Certain traders are more willing to take risks or drive hard bargains - this was reflected in the market price. Extraverted players are more likely to wander between country representatives, comparing different offers, thus having a better ability to find the best deal. They are particularly important for initiating debate over prices and quantities, and may also be more willing to propose alternatives when disagreements appear. Nevertheless, the strong, silent type can drive a hard bargain by his/her appearance of not shifting from the offer. Thus, both extraverted and introverted players (or countries) can bargain to their strengths, making for a potentially interesting outcome. Crucial to the success of these markets is an adequate supply of information. After analysing the marginal cost of abatement curve, the auction round provides information on the range of prices that might be paid. It is played with a degree of confusion, as some players are still not yet clear of their objectives and how to achieve them. The board acts as a vital source of information. It tells players what trades have taken place, which countries are buyers and which are sellers, and the current price for a permit. Producing a graph on the board as the trades take place recorded can be effective for helping players detect trends. In a couple games, I experimented with the supply of information. When each trade was made, the price was called out. We experienced substantially more activity and more stable prices than when information was less freely available, even though the only cost of the information to players was to go to the front of the classroom and examine the board. Just like real markets, there are other ways of gathering information. Espionage, sneaking over other teams’ shoulders, is popular and can help gather information about abatement curves, need for permits and prices willing to concede. Also, the players can consult outsiders (in this case, the lecturer) for advice on whether to agree a trade. No consultancy fees were paid.
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
S151
4. Conclusion This paper introduces the Carbon Trading Game and some of the observation from running it with students. The objective of the ‘role-playing’ game is to develop a feeling for how individual countries (and, thus, their companies) would behave when faced with commitments to the Kyoto Protocol and the possibility of tradable permits to comply with them. The game allows students to act as the leader of a nation involved in buying or selling tradable permits for carbon dioxide emissions to ensure they match emissions with permits at least cost. There are 10 countries (including Brazil, Canada, China, the EU, India, Indonesia, Japan, Russia, the Ukraine and the USA) with different costs of reducing emissions. The students are presented with their country’s marginal costs of abatement curves and total abatement curves. Students then need to collect information about potential prices per tonne to assess the trades (buying or selling permits) they will pursue. The game works very well at teaching the value of tradable permit schemes and some of their complexities, as well as being lots of fun. Another purpose of presenting this game is as an exercise in observing the nature and evolution of tradable permit markets for carbon dioxide. Each game generates interesting information about the nature of the market. The level of trade and prices vary substantially between games. The overall average trade volume was 321 million tonnes (or 6.7% of the total supply of permits) at an average price of US$ 19.5 per tonne of carbon for a total 6% Kyoto-style reduction target, and 499 million tonnes (or 9.2% of the total supply of permits) at an average price of US$ 41.9 per tonne of carbon for a 19.4% post-Kyoto-style reduction target. Running such experiments can also provide valuable lessons about the evolution of prices that might not be anticipated from economic theory. For example, the price agreed in the first trade has major implications for future trades. While there may be some adjustment towards the equilibrium price, the whole game is thrown in a particular direction. Also, the final flurry of trades often leads to an upward or downward spiral in prices. In addition, the evolution of prices depends crucially on the role of large players using their market power (Malik, 2002) and of technological innovations (Montero, 2002). Another important lesson is that this international permit system does appear to save industrialised countries money and earn developing nations highly sought-after foreign revenue (Stevens and Rose, 2002). While over-dependence on a market that could collapse—if, say, new technologies provide cheap ways of reducing carbon dioxide emissions—would be careless, these experiments remind us that many countries can gain from the creation of a market where the commodity traded is the responsibility for reducing pollution. Clearly, a simplified game representing a tradable permit scheme cannot do justice to the complexities of a real international market. For reviews of the market and institutional developments associated with an international tradable permits for carbon dioxide see Baron (2001) and Christiansen and Wettestad (2003). Nevertheless, the lessons from the games played can be of relevance to real carbon permit markets. An understanding of the nature of different countries (and their companies)—whether they are major or minor players, pro-active or reactive, informed or confused—will have important implications for how they interact and trade. The Carbon Trading Game will hopefully help many become more aware of potential gains from being involved in the burgeoning international market for tradable permits. It may also help some more vulnerable groups avoid the pitfalls of an international market for the right to emit carbon dioxide.
S152
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
Acknowledgements I would like to thank many enthusiastic students who participated in the game (mostly from the Imperial College MSc in environmental technology, the University of Surrey MSc in environmental engineering and members of the Foreign and Commonwealth Office) and Martin Hession, Matt Leach, Peter Pearson, Tanya O’Garra, Suzy Hodgson and Tim Jackson for their encouragement in developing and drumming-up interest in the game. I am also grateful to Denny Ellerman, Hank Jacoby and Annelene Decaux for allowing me to use their data for the purpose of the game. Appendix A. Example of the game The following is an example of the hand-out for country A:
A.1. Tradable permits game (country A): Students will have an opportunity to test their understanding of how tradable permits systems achieve the desired level of pollution and reduce the costs of abating pollution. They will play a game simulating a worldwide tradable permits market for carbon dioxide emissions. The following text would be form the background for the person (or persons) representing country A.
A.2. Context of the discussion: You (and your colleague) are representing country A. Your current carbon dioxide emissions are at current emissions 100 million tonnes per year and you have agreed to reduce them to 90. This means that you have 90 permits each allowing you to pollute 1 million tonnes Fig. 3. A tradable permit scheme has been agreed upon, which means you can buy a number of permits from other countries at a price in US$ per permit (such that you are entitled to pollute an additional amount equal to the number of permits) or you can sell a number of permits to other countries (such that you are entitled to emit carbon dioxide according to your new amount). Anyone failing to comply with emissions permits will be fined US$ 200 million per 1 million tonnes over your permitted amount. With this agreement in mind and the knowledge that you have recently been elected, you are trying to do ‘what is best for the country’.
Brazil
300
$ per tonne
Marginal Cost of Abatement
400
200
100
0
76
80
84
88
92
CO2 Emissions Fig. 3. Marginal cost of abatement.
96
100
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
S153
Brazil
3000
Million $
Total Cost of Abatement
4000
2000
1000
0
76
80
84
88
92
96
100
CO2 Emissions Fig. 4. Total cost of abatement. Source: figures based on Fig. A.4 of A. Denny Ellerman, Henry D. Jacoby and Annelene Decaux (1998) The effects on developing countries of the Kyoto Protocol and CO2 emissions trading. Working Paper 41. Joint Program on the Science and Policy of Global Change. MIT (available on webpage: http://www.web.mit.edu/globalchange/www/rpt41.html).
A.3. Your objective: Your objective is to buy or sell emissions with other countries in an attempt to keep the costs of reducing emissions as low as possible. At the end of each year, we will assess your actual total costs of reducing emissions (based on the emissions reduced, the costs of reducing these emissions, the earnings from selling permits and the costs of buying permits) relative to your initial total costs of abatement (represented in Fig. 4).
A.4. Trading: 1. In Year 1, to start the game, and to provide information about prices, an auction will take place, where permits will be allocated to the four highest bidders. Please write how much your country is willing to pay per permit. The four highest bidders will receive 2% of their current emissions. Thus, paying that price for 2% and increasing permitted emissions by 2%. 2. Calculate the total cost of abatement, according to your new or unchanged permit level (based on the emissions reduced, the cost of reducing these emissions, the earnings from selling permits and the costs of buying permits). 3. The total costs of abatement will be written in a log book. 4. In subsequent years, there will be no auction, just trading. 5. You have X minutes, to agree trades with other countries. You can agree as many sales or buys as you want, provided you have permits available to sell. 6. When a trade has been agreed, the two countries need to write in the long book the quantity of permits and the price per permit traded, and the countries involved. 7. Calculate the total cost of abatement, according to your new or unchanged permit level (based on the emissions reduced, the cost of reducing these emissions, the earnings from selling permits and the costs of buying permits).
Appendix B. Data for the countries The marginal costs of abatement for the 10 countries were approximated from the equation provided in Table A.1 of Ellerman et al. (1998). The total cost is calculated as the sum of all the marginal costs. For example, the total cost for Brazil reducing emissions down to 94 is equal to the sum of the marginal cost at 98 million tons (i.e. US$ 20/tonne) times two million, at 96 million tonnes (i.e. US$ 40/tonne) times two million and at 94 million tonnes (i.e. US$ 60/tonne) times two million; 2 million × 20 + 2 million × 40 + 2 million × 60 = US$ 240 millions. Note that the units of measurement are metric tons of carbon.
S154
Table A.1 Marginal and total costs of abatement at specified levels of emissions (in million tons of carbon) for the 10 countries in the Carbon Trading Game Brazil Emission
TCA (million $/tonne)
400 340 280 220 180 150 120 100 80 60 40 20 0
3980 3180 2500 1940 1500 1140 840 600 400 240 120 40 0
60 56 52 48 44 40 36 32 28 24 20 16 13 10 7 5 4 3 2 1 0
14028 12348 10780 9324 7980 6748 5628 4620 3724 2940 2268 1708 1260 896 616 420 280 168 84 28 0
China 840 868 896 924 952 980 1008 1036 1064 1092 1120 1148 1176 1204 1232 1260 1288 1316 1344 1372 1400
Emission 70 72 74 76 78 10 82 84 86 88 90 92 94 96 98 100
EU MCA (US$/tonne)
TCA (million $/tonne)
300 240 200 160 130 100 80 70 60 50 40 30 20 10 5 0
2990 2390 1910 1510 1190 930 730 570 430 310 210 130 70 30 10 0
90 80 70 60 55 50 42 36 30 25 20 16 13 10 7 5 4 3 2 1 0
12380 10580 8980 7580 6330 5280 4280 3440 2720 2120 1620 1220 900 640 440 300 200 120 60 20 0
India 600 620 640 660 680 700 720 740 760 780 800 820 840 860 880 900 920 940 960 980 1000
Japan
Emission
MCA ($/tonne)
TCA (million $/tonne)
560 576 592 608 624 640 656 672 688 704 720 736 752 768 784 800
400 320 260 220 180 140 120 100 80 60 50 40 30 20 10 0
32480 26080 20960 16800 13280 10400 8160 6240 4640 3360 2400 1600 960 480 160 0
150 125 110 100 90 80 70 60 50 40 30 26 22 18 88 10 8 6 4 2 0
2030 1730 1480 1260 1060 880 720 580 460 360 280 220 168 124 14 60 40 24 12 4 0
Indonesia 60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 100
Emission
234 240 246 252 258 264 270 276 282 288 294 300
US MCA (US$/tonne)
TCA (million $/tonne)
400 340 260 200 160 120 100 80 60 40 20 0
10680 8210 6240 4680 3430 2520 1800 1200 720 360 120 0
200 180 160 140 125 110 95 80 70 60 50 45 40 35 30 25 20 15 10 5 0
14950 12950 11150 9550 8150 6900 5800 4850 4050 3350 2750 2250 1800 1400 1050 750 500 300 150 50 0
Russia 300 310 320 330 340 350 360 370 380 390 400 410 420 430 440 450 460 470 430 490 500
Emission
MCA (US$/tonne)
TCA (million $/tonne)
980 1008 1036 1064 1092 1120 1148 1176 1204 1232 1260 1288 1316 1344 1372 1400
200 160 120 100 80 60 50 40 35 30 25 20 15 10 5 0
26600 21000 16520 13160 10360 8120 6440 5040 3920 2940 2100 1400 840 420 140 0
220 200 180 160 140 120 100 80 70 60 50 45 40 35 30 25 20 15 10 5 0
3210 2770 2370 2010 1690 1410 1170 970 810 670 550 450 360 280 210 150 100 60 30 10 0
Ukraine 60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 100
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
76 78 80 82 84 86 88 90 92 94 96 98 100
Canada MCA (US$/tonne)
R. Fouquet / Climate Policy 3S2 (2003) S143–S155
S155
References Baron, R., 2001. International emissions trading: from concept to reality. International Energy Agency, Paris. Braaddbaart, O., 1998. American bias in environmental economics: industrial pollution abatement and incentives versus regulation. Environ. Politics 7 (2), 134–152. Christiansen, A.C., Wettestad, J., 2003. The EU as a frontrunner on greenhouse gas emissions trading: how did it happen and will the EU succeed? Climate Policy 3 (1), 3–18. Ellerman, A.D., Jacoby, H.D., Decaux, A., 1998. The effects on developing countries of the Kyoto Protocol and CO2 emissions trading. Working Paper 41. Joint Program on the Science and Policy of Global Change, MIT. (available on: http://web.mit.edu/globalchange/www/rpt41.html). Malik, A.S., 2002. Further results on permit markets with market power and cheating. J. Environ. Econ. Manag. 44 (3), 371–390. Montero, J.P., 2002. Permits, Standards, and technology innovation. J. Environ. Econ. Manag. 44 (1), 23–44. Stevens, B., Rose, A., 2002. A dynamic analysis of the marketable permits approach to global warming policy: a comparison of spatial and temporal flexibility. J. Environ. Econ. Manage. 44 (1), 45–69. Tietenberg, T.H., 1990. Economic instruments for environmental regulation. Oxford Rev. Econ. Policy 6 (1), 17–33. US EPA, 2003. ‘Monthly average price of sulfur dioxide allowances.’ (available on: http://www.epa.gov/AIRMARKET/trading/ so2market/prices.html).
Further reading Schmalensee, R., Joskow, P.L., Ellerman, A.D., Montero, J.P., Bailey, E.M., 1998. An interim evaluation of the sulfur dioxide emissions trading. J. Econ. Perspect. 12 (3), 53–68. Stavins, R.N., 1998. What can we learn from the grand policy experiment? Lessons from SO2 allowance trading. J. Econ. Perspect. 12 (3), 69–88.