The market is betting on climate change
AMERICAN PUBLIC opinion has been catching up to the scientific consensus on climate change, but still has a long way to go. In a poll conducted by the Pew Research Centre in 2016, just 48% of respondents said they believed that the Earth was warming because of human activity. One likely reason for such widespread ignorance is that most people do not incur any penalty for being wrong. In contrast, the group most likely to learn the facts are traders in financial markets based on climate data, who stand to profit if they understand the issue and lose money if they dont. The consensus view of such expertsas measured by the prices at which they buy or sell futures contractslines up almost perfectly with leading academic research. A new paper by Wolfram Schlenker and Charles Taylor of Columbia University studied the relationship between mathematical climate models and prices for contracts at the Chicago Mercantile Exchange, which have allowed traders to place bets on the monthly total of days that are hotter or colder than average since 2001. The authors assessed the match between the two sources predictions. The yardstick they used was heating and cooling degree-days per season. A heating degree-day is a measure of the amount of heating people are likely to want to compensate for cold weather. It is calculated by taking a nice comfortable temperature, 65 degrees Fahrenheit (18 degrees Celsius), and subtracting the days actual temperature. Similarly, a cooling degree-day is a measure of how much cooling would be needed to make a hot day bearable: if it is 80 degrees outside, that counts as 15 cooling degree-days. The researchers summed up each methods yearly prediction for cooling degree-days between July and September, and for heating degree-days between November and March. They found that the markets predictions were remarkably well-synchronised with the forecasts made by NASAs climate scientists. The mathematical models explained fully 94% of the variance in aggregate market prices. In many years, bettors barely deviated at all from the scientific predictions. For example, in 2010, they expected 8.0 more cooling-degree days than the annual average from 2002-18, almost the exact same figure as the 8.2 calculated by NASA. The relationship was weaker during winters than summers, but still robust. It is hardly surprising that traders make use of scientific models when making investment decisions. Real money is at stake. What is noteworthy, however, is that the correlation between the two sets of forecasts is so close that it suggests that they base their choices almost exclusively on these models. That does not mean that statisticians have perfected climate forecasting: in fact, the errors their models produce are still large. But it does mean that they appear to have incorporated into their predictions all the information that investors use to bet on the climate.