My previous post was meant to be a prelude to a point I want to make now. The strategy that we need to get out of a crisis is very different from the strategy that we need to adopt to prevent the next crisis. When Mumbai got flooded, we could depend on the goodwill of its citizens to rescue it. While planning to prevent the next flood, we will have to contend with the selfishness of the same citizens. Impractical capitalist that I am, I am thinking of ways to make use of this selfishness, rather than just rail against it.
We know that the Met department is hopeless. It has an incredible 65% error rate, which means that their predictions are no better than an astrologer’s. I have long been convinced that no one in the department does any actual work. They just have a program that pulls together random phrases and gives out a forecast. We are so accustomed to bad predictions that we are surprised that people in other parts of the world actually rely on their weather bureau forecasts. The Indian Met department’s response to these examples is that they actually have a pretty good model, but the Indian weather is so damn unpredictable.
There is a way to find out. Virtually everyone depends on the weather- Companies, investors, farmers. It should be possible to finance a private Institute for Weather Research out of funds endowed by private bodies. It may take a few years for the institute to get a realistic model of India’s climate up and running and then it will take a few more before the forecasts it makes start getting trusted, but I think that it is well worth the bet. Most of those who endow it will probably get as good returns on their money as from any other investment.
Another instrument that will lead to better weather prediction is Weather Derivatives, i.e. betting on the rains.
Many of you might be familiar with Commodity Futures. I bet with you on the price of a particular commodity, say wheat. If I sell you a wheat future at 1000 rupees a quintal, It means that we check the price of a quintal of wheat at an agreed upon date. If the price is above 1,000, you pay me the difference. If the price is below 1,000, I pay you the difference. This trade, if carried out publicly, gives everyone an incentive to predict the price of wheat accurately. A wheat farmer, by selling wheat futures, can protect himself against uncertainty in the market.
Weather futures is a similar concept. If a farmer needs 1000 mm of rain in the month of July to get a good crop, he can “sell” weather futures worth 1000 mm. If the rains fall below 1000 mm, he gets money corresponding to the difference, and to that extent he insulates himself against a crop failure due to insufficient rains.
If this trading is carried out publicly, it will enable better weather prediction. It will also result in more donations and consultancy projects for the Institute of Weather Research.
This, assuming that precise weather prediction is actually possible. And the only reason weather prediction is bad is because there is no incentive for it to be correct is it? Hmm…!
It depends on what you mean by “precise”. I’d rather use “dependable”.
Dependable weather prediction is in fact possible. I don’t know about the UK (you tell me) but in the US, people actually rely on the predictions for the next 48 hours. We are talking of “snow in the afternoon, heavy rain in late evening” kind of prediction. It might be wrong occasionally, but it is right often enough for people to plan their week. I don’t know if it is possible in Indian weather. I am just saying that we should find out.
The other kind of prediction is the forecasting for the season. I don’t want them to predict in March that Mumbai will be washed out on July 26, but atleast they should get “20% above average” kind of predictions right within a margin of error. I understand even this is possible in the US, I am saying we should find out.
Finally, if you are talking of weather derivatives, even if it is impossible to predict accurately, they still provide a way of hedging against the risk of a bad monsoon.
Ravi,
The conditions in the US and in India are very very different.
So far no one has succeded in “mapping the monsoon”. As a result, there simply is not enough science as yet to achieve the level of accuracy/dependability in India that the US weather stations have.
Possibly. But if it’s our weather department that is saying this, I’d like a second opinion. But still, is there any reason why tomorrow’s weather can’t be predicted? Surely weather prediction can’t be that impossible?
“Surely weather prediction can’t be that impossible?”
Certainly not impossible. Amma can 😉
I know of atleast one small firm in Mumbai called WeatherRisk which is involved in exactly the same derivatives you are talking about. Maybe you can check it out. Some banks like ABN AMRO are also getting into this business.
Sandy do you have a URL I can check out? Searching for WeatherRisk Mumbai points back to this page 🙂
Of course, I am not claiming that I thought up this concept of Weather derivatives. It already exists. I just didn’t know that it is allowed in India.