Island News & Views
Go to Site Index See "Island News & Views" main page
Expired · 23rd February 2011
Ray Grigg
Predicting the future is fraught with inaccuracy. Even experts are known for their remarkable inability to anticipate the results of unfolding circumstances. But a professor of politics from New York University has become a rare exception.

Over a 30 year period, Dr. Bruce Bueno de Mesquita has an average of 90% accuracy for thousands of predictions – he doesn't predict random events such as lottery draws or issues involving millions of variables such as stock market prices, financial crises or general elections. But he is incredibly accurate at predicting outcomes where a limited number of people interact with "negotiation or coercion, cooperation or bullying" (New Scientist, Mar. 20/10). This includes geopolitical and strategic situations pertaining to "domestic politics, foreign policy, conflicts, business decisions and social interactions".

Bueno de Mesquita accomplishes the nearly-impossible by using game theory and mathematical models to assess what people will do in situations that depend on other people's decisions. His operating principle is "that people do what they believe is in their best interests". The outcome is determined by the complex interaction of all the self-interested reactions of all the people involved – five people generate 120 interactions while ten people generate 3.6 million interactions.

Since "garbage in equals garbage out", the gathering of accurate data is extremely important. From an arbitrary scale – 1 to 100, for example – a number is assigned for each person's involvement in each of four categories: the outcome desired, the importance of the issue, the determination to reach an agreement, and the weight of each person's influence. A computer algorithm processes the numbers and arrives at a number that indicates an outcome on a graduated scale between two defined extremes. The numerical outcome must be interpreted but Bueno de Mesquita says it is not ambiguous.

In one example, on the question of whether or not Iran would build a nuclear weapon, an outcome number of 120 on a scale of 1 to 200 indicates that Iran would enrich weapons-grade uranium but would not build a bomb. In another example involving foreign aid to Pakistan, the American government wanted to know how much foreign aid would be needed to convince Pakistanis it was in their best interests to eliminate terrorists within their country. The model predicted that aid of $750 million would have to be doubled. But even with $1.5 billion per year, Pakistan would only pursue terrorists on a scale of 80 out of 100 because the total elimination of terrorists would dry up their American foreign aid. In this particular assessment, self-interest would trump peace.

In a related but more important matter, Bueno de Mesquita ran his calculations to predict the outcome in 2050 of international negotiations to reduce greenhouse gas emissions and control global climate change. His prediction is not promising. The model expects that stringent reduction targets will be set, but not met, as countries such as Brazil, India and China struggle to rise in political and economic power. Canada has already established precedent in this regard by committing to targets and then abandoning them.

Although we still have 39 years until 2050, recent events are supporting Bueno de Mesquita's prediction. The definite targets set in 1997 by the Kyoto Protocol have not been met – emissions have gone up rather than down. Negotiations at Copenhagen in December of 2009 backed away from these targets. And the 2010 discussions at Cancun, Mexico, were only a success because they were not an abject failure – optimists argue that foundation agreements were reached that will accommodate the major decisions that must be made in 2011 at Durban, South Africa.

Nature, of course, responds to reality, not intentions. Negotiation is not action. And the longer negotiations last, the more dramatic must be future corrective measures. The Cancun meeting agreed that the target for the highest global average temperature increase should be 1.5°C, with the proviso that 2.0°C should be the absolute ceiling. Research from the Climate Action Tracker project, however, indicates that existing pledges to reduce greenhouse gas emissions will set the temperature increase at 3.2°C (Guardian Weekly, Dec. 17/10).

If Bueno de Mesquita is correct that people act on the basis of their "best interests", at what point does preventing global climate change become a common priority? Such a question is complicated by the fact that "best interests" are not distributed equally around the planet. This partially explains why Canada has been both a laggard and obstructionist in United Nations' climate negotiations. While rising oceans are submerging South Pacific islands, while Africa's Sahel is cooking in drought and Australia is drowning in flood – Australians had their drought last year – Canada has not been motivated by dramatic climate disasters. Indeed, unlike the turmoil that will be generated in much of the rest of the world, Canada may even experience a net beneficial effect from global warming. Meanwhile China, India and Brazil are too busy with economic growth to risk serious carbon dioxide reductions.

Whether this situation lasts until 2050, as Dr. Bueno de Mesquita predicted, remains to be seen. But solving a global problem without a global agreement almost certainly subverts any solution. The path our leaders are now choosing, however, offers no return. Once started, the ecological and climate changes that are being set in motion cannot be stopped within a time frame meaningful to our present civilization.

If "forewarned is forearmed", then knowing the future allows us to change it. This is the sobering relevance of Bruce Bueno de Mesquita's predictions. With an accuracy rate of 90%, we won't be able to complain that we weren't warned.