26 February 2013

The danger of flashing wrong signals

Have you heard the stories of people who have driven through fields, into lakes or off cliffs while following their GPS units? Any outsider would have told them to use their common sense before making a right turn over a cliff, but are WE so wise when it comes to our indicators?

Although the water sector really needs more and better information (that's why I founded the water data hub*), I worry about people putting the wrong weight on the wrong information -- a worry that puts these recent stories into a different context:Now, I'm not worried about the discussion of water risk. I think that it's a topic of growing importance, as the end of abundance exposes business models, bureaucratic assumptions and personal habits formed in an era of too much, too cheap water to a new reality of scarce water that cannot be taken for granted.

What I worry about is the arrival of over-simplified models of risk and tools for "managing" risk, and the damage that may result from the naive application or use of these models.

I worry, for example, of what might happen if:
  • Some 25 year-old bets your pension based on "data" from his Bloomberg terminal (Bloomberg partners with WRI).
  • Banks or governments allocate investments based on the IBM-model of risk.
  • Investors move their money based on Ceres' risk warnings.
All of these actions may be valid in their consideration of the information at hand, but what if they are paying too much attention to what's in the index while missing what's NOT in it?

I'd put a lot more weight on:
  • Local institutions for managing water. They determine the difference between drought and shortage.
  • Political and regulatory interventions (i.e., "regulatory risk") that outweigh natural risk by an order of magnitude.
  • The importance of funding local water projects with local money -- especially when it comes to knowing the people who are spending your money.

Each of these factors, IMO, are significant, and each can overwhelm any number that you get from a risk model.

Bottom Line: Caveat emptor! Risk in the water world is not so much about nature or market forces, but the arbitrary, illogical, unaccountable and uncertain ideas and actions of politicians and bureaucrats who drive water management (and cause 90 percent of the mismanagement).
* The WDH does not worry me because it's merely a way of finding data (is yours linked?). People can then interpret the data any way they want.

H/Ts to BB

5 comments:

Hyun sung Ha said...

Good point, David. Makinging an open database that is both popular and reliable would be best.

niksin said...

Thanks for posting this; this is a relevant trend that is likely going to control the water debates for a while. At the moment there are more organizations working on water risk tools like these, and from my perspective they seem to be in competition over access to funds, media exposure etcetera.
It is much like the "water footprint" discussion or even the "water stewardship" approach. Over the years, I have been in contact with several of the people behind these ideas; though they do have good intentions and some good ideas, in general they do not acknowledge the full complexity of water issues, or the weaknesses of presenting their tools and interpretation as a final risk assessment.
I have recently started on the idea (WaterLeaks) to create a transparent website that collects all these different water approaches and let people write critical reviews on their experiences and suggestions (e.g. tripadvisor-style). Map out the organizations and sponsors behind the tools, and probably allow people to add case studies. On current capacity, it would take months before anything public and interactive comes out. As far as I know, such a transparent meta-level on water tools is not out there yet... anyone please let me know if it already is, would save me a lot of work.

David Zetland said...

@Niksin -- Agreed. I've a plan to have such "comments on quality" on the water data hub in the future...

AP said...

It seems, everyone likes to join the water risk bandwagon. Have to admit, I was one of them when I developed XXX and currently expanding the same by bringing in energy and agriculture components to it.

I feel, any model which tells the story using global data would almost always make hollow claims with reference to (i) time, and (ii) resolution. Most of global water data (e.g. risk, demand, supply) currently used are circa 2000. Secondly, water being a local issue one needs local (higher resolution) data. For example, even if one has storage capacity of dams, when and how much water is released is equally important for demand/supply and risks aspects.

I think waterfund, Bloomberg (has tied up with SIWI to develop yet another index, not sure if they’ve still continued their association with WRI!) are simply trying to define rules of the game especially among investor community; once done they can provide manuals. But I think, they still have a long way to go. We’ll see!

The trouble is there are too many non-water issues responsible for the current state of water. Energy availability is a case in point. For me water is the most assuredly renewable resource on the earth. Energy is not. If energy and materials are not limiting factors, I can desalinate Indian Ocean and pump it to Mount Everest. In Africa, for example, if one resolves access to energy issue, there is plenty of groundwater, one can easily solve drinking water problem. So to solve water problem at least n Africa, one needs to invest in energy first. The vicious circle for water in fact starts there.

Seth said...

I liked this post, and I agree that there are clear dangers in having (mostly opaque) metrics that result in the movement of huge sums of money.

Incidentally, one thing that continues to bother me is that most of these risk models--Aquaduct included--ultimately rely on the (largely arbitrary) categories of water stress that were outlined by the United Nation's Comprehensive Assessment of the Freshwater Resources of the World (1997).

In this report, watersheds with demand/supply ratios ("criticality" ratios) of >0.4 are said to be experiencing medium-high water stress. 0.4 is the magic cut-off that always seems to appear at the end of the chain of citations. It's eerie!

Check out the UN text that was posted here, and specifically 71(c): http://jzjz.tripod.com/freshwat.html