Research
Input-Output models describe an economy as a web of mutual supplies: every sector’s output is another sector’s input, so a change in one place is never confined to it. I work with the environmentally extended version of that framework, in which a physical quantity - water above all - travels alongside the monetary flows. That makes it possible to ask how much of a resource an activity uses directly and how much it draws through everything it buys, to decompose a change in resource use into the forces behind it, and to follow a disruption as it spreads through the structure of production. The computation is done in R, on national supply-and-use and Input-Output tables combined with environmental accounts. The reason for working this way is that the damage from a physical shortage is mostly indirect, and indirect damage tends to go uncounted: it is assessed once the season is over, sector by sector, and whatever cannot be traced drops out of the total. A model of the dependencies makes that part visible in advance. The longer-term aim is a usable tool: a model in which a disruption can be simulated and mitigation strategies compared before a drought, rather than accounted for after one.