The Microeconomics of Climate VaR
Programming note: Our apologies for delaying the Wednesday post this week, scheduling issues prevented its timely publication. Louie will publish his weekly article Saturday (tomorrow)
When you first studied microeconomics, I’m sure you found the distinction between the long- and short-term very confusing. I know I did. In the model of perfect competition, assets are assumed to be fixed in the short-term, which means that the costs of ownership cannot be avoided. In the long-term, asset owners have various options regarding the handling of fixed assets, which means that any and all fixed costs become variable.
It still makes my palms sweat a little – I only just passed micro the first time I studied it, though that probably had something to do with the fact that I was spending too much time in the uni bar.
Anyway, the point is that the long- and short-term behavior of economic agents can be very distinct. Any micro analysis of the long term must factor in the various options that asset owners can take as circumstances change and the value of their assets wax and wane.
I was reminded of this distinction recently when pondering Value-at-Risk (VaR) in the context of climate change. VaR is defined as the size of financial loss that may be expected over a certain time frame at a given confidence interval. It is a widely used measure of financial risk that will be very familiar to most readers working in the banking and insurance industries.
The seminal paper on Climate VaR was published by Dietz, Bowen, Dixon and Gradwell in Nature in 2016. In it they used simulations from an extended version of Nordhaus’ DICE model to compute the VaR for a representative portfolio of financial assets. Here’s how Dietz et al (2016) describe their key results:
“Table 1 provides estimates of the impact of climate change over the course of this century on the present value of global financial assets. Along the DICE baseline or business-as-usual (BAU) emissions scenario, in which the expected increase in the global mean temperature in 2100, relative to preindustrial, is about 2.5°C, the expected climate VaR of global financial assets today is 1.8%. As Table 1 indicates, there is particularly significant tail risk attending to the climate VaR. The 95th percentile is 4.8% and the 99th percentile is 16.9%.”
The phrase that I’ve been puzzling over is “global financial assets today.” I don’t think they mean the Climate VaR of the precise set of financial assets that is currently held, because the work is based on a macroeconomic representative agent model. The authors’ calculation assumes that the representative investor lives forever and constantly rebalances their portfolio to precisely match the concentrations found across the entire economy.