Recipe for Failure

1. China and India may show higher support for an agreement with differentiated commitments from developed and developing countries.
2. The relative power of nay-actors will increase throughout the century, but that won't stop NGO pressure and concrete climate change impacts from influencing their position.
3. Wouldn't stagnation in emission controls be more likely than actual reversion?
4. After China and India, 1/3 of the human population, the next wave of emerging countries will be relatively small and more easily controlled. Can there still be hegemonic wars in the nuclear age?
5. Climate change will be a net benefit perhaps to countries around the arctic circle. That's it. All others will suffer losses from more extreme weather. There is no reason to expect a beneficial reallocation of renewable energy resources. These impacts will be a strong incentive for innovation - but wouldn't it be smarter to act on these incentives right now?

The model is quite simple, I guess. And it just works, like fuzzy logic, or google.
But - "New technologies will solve the problem for us" - just go and put a question "What kind of really new technology will appear and approx. when (in tens of years) it will appear?". Now, think about new cars, then about new windows system, then about new regulations in general, what kind of new technology we can expect? Regarding new technology, actually, look on Harley-and-Buell issue, or Oracle-and-Virtual-Iron issue. More - try to predict how the government's expenses on they're own living and its bureaucracy will behave.
I'd recommend Asimov's "Foundation", and Huxley's "The great new world".
Have a nice reading.. I really hope "new technology" appear, and even, someone like you - or me - will invent it. I have doubts, unfortunately.. anyway, giving up is not the strategy.
Qui vivre, verra, I guess ;-)

So what we learn is that a little bit of theoretical rigor- of the sort most social sciences bring- is a useful way to structure an enormous amount of data! We learn that knowing the preferences and chosen strategies of a large number of actors, indeed knowing who the relevant actors are, makes it much easier to work out what will happen. Um, yeah.

That's no manifesto for game theory, or BdM's charming attempt to produce superfreakpoliticalscience. It's called social science: know your stuff and then use theoretical tools to say meaningful, generalizable things. BdM's chosen form of theoretical tool is game theory, which brings costs and benefits (the benefits are in analyses of decisions, which is why it's useful in thinking about decisions). He also happens to be quite good at the data collection, which is why he gets useful places with game theory that is not all that sophisticated.

The economists- not to mention a horrible virus that was in the water in Rochester a few decades back- have taught us that this operation (called social science) must be done with enormous hubris. They sound like Comte in the history of sociology and will seem just as strange in a decade or two. But Walt was right: game theory has added very little that Max Weber didn't know.

I congratulate BdM on graduating from journalism to social science in his conversation with the State Department official, but he really shouldn't be trying to turn himself into the LTCM of politics. If nothing else, it undersells his own impressive interviewing skills and will produce many more misguided PhD students who buy the line that knowing the game theory beats knowing the facts.

..."and I can prove it" -- A computer model is not a mathematical proof.

Game theory proven mathematical fact, and as such it should be a welcome addition to the tools available to policy makers. I do not think however that it should replace local expertise. Someone has to put the numbers into the model, and no one knows the situation better that the "troops on the ground".

If Mr. Bueno de Mesquita really believes that his model will help alleviate global suffering, he should release the source code to his model for peer review. If such a thing really has benefit, then we should be putting it in the hands of every state department official.

With some modest web searching, I've found no links to Mr. Bueno de Mesquita's model. The results are typically fascinating but sensationalistic news stories like this one.

I think it is telling that you begin your process by grabbing a "yellow pad" and write down "how much clout they [the actors] had, what their preference was between the various plausible candidates for prime minister, and how much they cared about trying to shape that choice". So, the starting numbers for your computer model are totally subjective, even if the model itself is rigorous. It follows that your entire approach is based upon the (unstated) assumption that the initial data on your "yellow pad" is easier to quantify/estimate than final outcomes. local expertise is still vital to your process, but you use it in an unorthodox way.

Perhaps I am missing something, but I cannot find where you have included GLOBAL WARMING in your calculations. Do you believe that facts do not influence political decisions? Are the lives and needs (AND decisions) of the peoples of the earth factored into your calculations?

I assume you are aware that some scenarios (see for example http://arxiv.org/abs/0804.1126 ) developed by our best scientists include the possibility of sea level rises of meters in this century. Do you think the effects of global warming will not change how decision-makers act?

It appears the model is missing the fact that oil is a finite resource. Climate change aside, the world will have to move to other forms of energy eventually (depending on when one things peak oil has or will occur) due to lack of low-cost fossil fuels to burn. Sure, there will be plenty of coal still, but we can hardly run today's economy on coal.

The model is predicting the emissions standards in international treaties, but the entire question may well be irrelevant in fifty years.

How do we factor in the simulation the act of God?