## 16 Mar 2012

### Bleg: Statistics for water inputs to final products?

TS wanted to do a "silicon chip to potato chip" comparison of the output value per unit of water.

I cautioned him that we cannot just divide the TOTAL value of final products by the amount of water consumed. We need to measure the MARGINAL value of water to the output, i.e., removing the effects of other inputs, labor, capital, etc.

Does anyone know of statistics that quantify the value added from water across many industries or products?

Mr. Kurtz said...

Since there is no set of generally agreed-upon rules, the studies I have seen that try to measure this basically just support whatever the bias of the experimenter was. "One hamburger takes eighty eight zillion gallons of water", that sort of thing; even more if you count the showers the cowboy had to take. A best selling novel, or a masterpiece painting take almost no water to produce. It's kind of beside the point.
You are right about the marginal cost. If water is priced appropriately, rational people will do things with it that creates the most economic value. Transportation and storage is a major cost people overlook when comparing silicon chips and potato chips. Most agricultural water is far removed from users who might turn it into something more valuable than food. By the time an industrial user builds a delivery, treatment and storage system (because, unlike farmers, they need water 24/7/365) that "cheap" agricultural water might not be so cheap after all.

mac said...

Isnt such a comparison impossible without weighting the volume of water used by some 'local scarcity' factor and perhaps by some 'water quality' factor?

Veolia's index tries to incorporate these factors (which is better, logically-- but i cant speak to how well this works in practice)

Assuming we could pull this off, at great cost, what would it all show? Exactly what DZ says: marginal value would = marginal cost. That in water scarce or sensitive places only a high economic value per unit is justified.

So i guess the better question would be how to price water better, right?

Carlos A López Morales said...

This sounds familiar to the sort of measurements you can do using Input-Output techniques, but I'm not sure those are the ones you're looking for...

In particular, I'm thinking on this study:
http://upi-yptk.ac.id/Ekonomi/Velazquez_An-input.pdf

Also note there's an special issue of Economic Systems Research on Input-Output and water (I have one paper on Mexico in that issue :o))
http://www.tandfonline.com/toc/cesr20/23/4

I've seen other calculations. For instance, taking water consumed in a given sector and divide it by the value added in that sector. I don't think this is very informative for reasons similar to those you say...