Wednesday, 4 March 2015

Climate Change Debate Part I: Confusion Between Risk & Uncertainty



This is a preamble to my forthcoming series on climate change and green polices. I only want to make one key point here - that despite what the vast majority of people involved in these debates will tell you to the contrary - any future discussions about climate change polices are not primarily discussions about risk, they are primarily discussions about uncertainty. I'll explain.

Some in the green contingent say they are fairly certain of the negative effects on the climate in the forthcoming decades, which, according to them, justifies radical climate change policies in the here and now. Others say there is real uncertainty about how bad things are going to get in the future, which, also according to them, justifies climate change policies in the here and now. They can't have it both ways: they can't use both near-certainty and near-uncertainty to argue for a policy. Professor Lord Stern falls in the latter camp, holding the absurd idea that uncertainty should generate extra radicalisaton in the present - an idea he promulgated when he was adviser to the UK Government on climate change: "Uncertainty of impacts strengthens the argument for mitigation", he (in)famously said.

Professor Stern is wrong; uncertainty of the impact of climate change strengthens the argument for non-mitigation, not mitigation. This rule doesn't apply in every sense, of course - sometimes uncertainty should engender extra caution. If I can't remember if I locked the front door, it is a good idea to go back and check. If I have increased knowledge that burglars are on the prowl that day then that should increase my urgency to go back and check the door is locked. But unlike the situation with the front door, where I know that a locked door is more desirable than an unlocked door, I don't know if the costly and economically-inefficient climate change policies are more desirable than undesirable, because we lack the information to calculate the risks.

To not understand in the way that professor Stern is not understanding this is to mistake risk and uncertainty. Risk is assessing the potential costs with known probability. Uncertainty, on the other hand, is not knowing the probability, which means an inability to calculate a risk. If I have to draw a Jack, Queen or King card from a 52 card deck to win £1,000,000 or else die, that is a ‘risk’ because I can calculate the probability (12 in 52). On the other hand, if I have to draw a Jack, Queen or King card from an unspecified pile of cards, and I don't know how many are missing from the pack, then I have ‘uncertainty’. I cannot calculate the probability of drawing a picture card because I don't know if any picture cards have been removed.

Let me make it even clearer with an illustration. Suppose there is a pile of 99 cards - all of which are either a Jack, a Queen or a King, and all three cards are represented. You know that 33 of the cards are Jacks, but you don't know the ratio of Queens and Kings in the remaining 66 cards. You can choose from two scenarios:

Scenario 1: You win £1,000,000 if you draw a Jack, and nothing if you draw a Queen or King.  

Scenario 2: You win £1,000,000 if you draw a Queen, and nothing if you draw a Jack or King.

Which scenario would you prefer? Due to scarcity of information there really is no way to know which scenario is preferable because you don't know the ratio of Queens and Kings - you only know there are 33 Jacks. If you choose Scenario 1 you know you have a 1 in 3 chance of £1,000,000. If you choose Scenario 2 you don't know what chance you have because you don't know how many of the remaining 66 cards are Queens - there could be as few as 1 or as many as 65. Scenario 1 offers you a risk; Scenario 2 offers you uncertainty.

The climate change assessments are generally more like Scenario 2 than Scenario 1 - they involve uncertainties where drastically little is understood about the probability. It was important to mention that before we got under way with the series. In the next part I will look at how mindful we should be of future generations, and what we owe them.

 
* Photo courtesy of datagenetics.com
/>