06 December 2011

Should water companies self-regulate their risks?

This white paper [PDF] from Ofwat proposes a move from output- to outcome-based regulation of water companies in England, i.e.,
The first change is to the current system of regulatory reporting. This would move Ofwat away from detailed regulatory monitoring of compliance. This change puts the companies firmly in charge of managing their risks, and redefines our role as holding them to account for the results, not the processes they adopt to ensure compliance.

Second, we propose that Ofwat itself adopts a risk-based framework. This will not only include a systematic assessment of risk and opportunity but will also allow us to focus the allocation of our own resources. This will demonstrate that we intend to conserve resources and target our efforts, so we have the capability to act swiftly and decisively where there is a real risk to outcomes for water consumers. And the companies and we would be freed from regulatory burden where there are no material risks to consumers.
This project, while admirable, suffers from three fatal flaws:

First is the emphasis on self-reporting and regulation of risk by companies that have little incentive to provide accurate reports on risk. Mistakes, after all, can be plausibly blamed on "unexpected, outside" factors beyond the companies' control, i.e., "can we increase our prices to fix the problem please?"

Second, "self regulation of risk" does not instill confidence in people after the fiasco of self regulation of risk by financial companies that have absorbed $ trillions in bail out money, world-wide.

Third, Ofwat faces information and incentive problems in terms of monitoring risk inside companies. What questions should be asked? Who wins or loses if the wrong questions are asked or the wrong weights (in terms of risk) are applied to the answers? How will outsiders know if Ofwat is doing a good job at monitoring one company, let alone 20+ companies with very different institutional and physical characteristics.

Luckily, I have an answer to these flaws:
  1. Rebrand this effort as "delivering sustainable services" instead of "managing risk." 
  2. Require that water companies take out performance insurance with companies that will compete for business (minimizing the cost of insurance), watch water companies (providing oversight), pay for accidents (relieving customers of higher bills for mistakes), and provide a transparent view on performance (facilitating comparisons and benchmarking). All of these ideas are spelled out in this paper.
Bottom Line: It's difficult to balance between over-investment that taxes customers and under-investment that puts their water services at risk, but regulators can facilitate that balance by importing a competitive industry that will take responsibility for failure while getting paid for success.

5 comments:

Mohamad Mova Al' Afghani said...

rebranding does not the incentive problem, david...

i think one of the incentives for making good reports are that companies would be put to more scrutiny for under reporting.

my phd touches upon this issue too. my question is: will the move towards risk based regulation decrease the availablility of information in the public domain? note that ofwat will no longer require annual june return submission. the june return is published. without any collection, then there will be no more information in the public domain (except for the obligatory publication of utilities regulatory account. what do you think david?

Eric said...

I agree with your critiques. I think that your solutions also have fatal flaws.

For example, where do the competing insurers come from? Why are they in the game? Who pays for the watchers of the water companies? If the payer of the watchers is the government and the politicians stand for reelection every two years or when a crisis of confidence occurs, doesn't this mean that there is no accountability of the watchers? Who actually pays for the accidents? Is it the taxpayers? If the accident is twenty years in the making, who should pay what fraction of the costs? What does 'transparent view' mean in detail? If it means filling out more forms, which are never read, then the process is technically transparent but in fact opaque.

I look forward to your second round of answers.

David Zetland said...

@MMAA -- those reports were not only non-credible (who audited them? who paid if they were wrong?) but they were too voluminous to be accessible. Read Hayek to know why price signals (insurance premiums) will convery FAR MORE information in a few data points.

@Eric -- there are many insurance companies. It's easy for the regulator to make the "market" contingent on 10+ companies operating (how many companies NOW insure business risk?)/ They are in the game to make money. They (not govt) are the watchers so they are paid. Accidents are covered by insurance. Transparent is NOT from forms, but from the resulting premiums that are public information. You need to read my paper :)

Eric said...

Read the paper.
Need some financial details and discussion of conflicts of interest.

Eric said...

Some details that I need.

Can insurance companies refuse to insure water utilities, as has happened in automobiles and medical insurance?

What is the cost per year and terms of a typical policy?

Where are such policies in effect now?