An article at VoxEU, “Climate change negotiations PLC?” by Ralf Martin gives a nice recap of some prediction markets on climate change just launched at Intrade. I wish he had gone a bit more into the limitations of prediction markets; while they aren’t a panacea, they have their uses. Big caveats: the markets need to be reasonably deeply traded and the quality of results is dependent on how well informed the participants are, and whether they in aggregate have biases.
A new climate change prediction market has been launched. Here are the details on motivation and participation.
What will happen to global efforts to limit greenhouse gas emissions when the current commitment period of the Kyoto protocol ends in 2012? There are hopes that the United Nations (UN) climate change conference that started on 3 December in Bali will outline a road map for a successor agreement that might be reached in late 2009.
To know what is going to happen in 2012 and beyond is not only relevant for future generations but for a wide range of investment decisions today:
Should you buy that flat in a riverside building?
Should companies invest in R&D on carbon capture and storage technologies?
Should governments provide more money to wind farm development or rather improve flood defences and relocate people living close to the coast?
Should an energy intensive company relocate its production from an industrialised country to a developing country unlikely to be covered by a pollution target?
To make these choices wisely, we need forecasts on the likely outcome and design of any post-Kyoto agreement. There is no shortage of opinions on what is going to happen. But whom should you trust? Opinions will be influenced by wishful thinking and strategic positioning alike as well as being based on different insights into parts of the puzzle. For example, living in the rich world, you might have a good sense of what the typical voter in your country thinks about pollution targets, but what about Indian voters or the Chinese communist party?
If the UNFCC – the UN body in charge of the climate change negotiations – was a publicly listed company whose profits depended on the climate change reduction targets it achieves, a good way of getting an idea on what’s going to happen is to have a look at its share price rather than public statements of its bosses or shareholders. This is obviously not a possibility. But what is possible now might be even better.
Last week, Intrade.com, the Irish prediction market company, started trading in a range of financial contracts whose payoff depends on specific outcomes of a post-Kyoto climate change agreement. The contracts, which were designed together with the Centre for Economic Performance (CEP) at the London School of Economics, capture which countries will participate in a post-Kyoto agreement as well as how stringent any pollution reduction targets might be.
How does this work in practice?
All climate change contracts on the Intrade market are binary contracts. Consider, for example, the US.TARGET.DEC09>10% contract. If by December 2009 the UN negotiation process has led to an agreement that implies a 10% reduction target for the United States relative to 1990, this contract pays $10. This is good news if you bought such a contract for less than $10, because you have then made a profit.
Of course we do not know for certain today if there will be such a target for the United States. The only way for you to buy such a contract for less than $10 is because some other market participant believes the United States will not accept any agreement limiting their emissions.
If enough traders who are sufficiently informed participate in such a market, the realised prices are not only a reflection of the market participants’ beliefs but might also be good predictors of the actual outcome. In other words, if contracts trade at a price of $7, our prediction is that there is a 70% chance that the United States will accept a 10% or higher target.
Nobody has examined if this works for climate change related contracts. This is one of objectives of the current initiative. But there are encouraging results from other contexts such as elections…..
What about other climate change issues?
The outcome of the UN climate negotiations is of course only one of many uncertain factors relevant for making good policy and investment decisions related to climate change.
Other issues include the correctness of many of the scientific theories and assumptions on which climate and economic models are based, the scope for different technologies to mature and to be adopted, unilateral actions by various governments, behavioural changes by consumers, actions by businesses etc. Many of these can potentially be captured by prediction markets.
There are a number of factors that have to come together if prediction markets are usefully to be employed. These include:
There must be an event or state of the world which is uncertain before it happens but clearly observable for everybody involved at the end of a contract.
The outcome of the event must be contentious which usually happens if all relevant information to predict an outcome is dispersed with different people having access to different subsets of the complete information set.
Sufficiently many traders with access to relevant information subsets need to be attracted to participate in the market for whatever reason: they might be interested in profit opportunities, the subject matter or simply the thrill of gambling.
While it would be great if a climate change prediction market would provide useful forecasts, note that generalizing the success of election markets has its limitations. The people betting in those markets, for the most part, were voters (whether they voted or not) and a vote is a binary outcome (at least in America, with a two party system). Thus their actions in the prediction market were very similar to ones they’d take, or could take in the voting booth. In climate change, you will have experts and non-experts setting prices. The outcomes could well resemble the result of a poll, and be unduly influenced by efforts to manipulate public opinion. And there are more dollars in trying to sway opinion against the urgency of combating climate change than for it.
In fact, could you see, say Exxon Mobil gaming the market to prove the point that global warming is an overrated issue? The costs would be trivial compared to a national ad campaign.